Hi,

I fear the top-leader of the warnings (hp-option.c) will stay on the 
list. Most of the warnings are "unused parameters". This is because the 
functions are called through a single function pointer. And some of the 
functions really don't use the parameter. A solution could be to add 
global variables to which the unused parameters are assigned. This is 
what I dont't want to do.
Any other solution without adding dummy code just to satisfy the compiler ?

--Peter

Henning Meier-Geinitz schrieb:
> Hi,
> 
> This is a summary of the compilation warnings caused by gcc-3.0.4 on
> Linux. Some of them are rather harmless (e.g. unused parameter), but
> could be avoided nevertheless. Others (like comparison
> unsigned/signed) may be an implementation bug. Please check.
> 
> The following two tables list the numbers of warnings per file:
> 
> ./configure --disable-warnings:
>       2       sanei_scsi.c
>       2       sanei_usb.c
>       4       agfafocus.c
>       5       as6e.c
>       2       avision.c
>       8       canon-sane.c
>      23       coolscan.c
>       1       coolscan2.c
>       1       dc25.c
>      10       djpeg.c
>      12       dmc.c
>       2       epson_scsi.c
>       2       hp-accessor.c
>      53       hp-option.c
>       2       hp-scl.c
>       2       hp.c
>       6       microtek.c
>      25       mustek_pp.c
>      12       nec.c
>       3       net.c
>       4       pie.c
>       4       qcam.c
>       1       ricoh-scsi.c
>      12       s9036.c
>       8       sharp.c
>      14       sp15c.c
>       1       sp15c-scsi.h
>       1       sp15c.c
>      11       st400.c
>       4       tamarack.c
>      13       umax_pp_low.c
>       6       umax_pp.c
>       2       dll-preload.c
>       1       dll.c
> 
> Full log at:
> http://www.meier-geinitz.de/sane/tmp/warnings-2002-11-16-gcc3.0-nowarn.log
> 
> With all warnings enabled, there are some more "false alerts". E.g.
> the compiler complains about printf ("%p", pointer), where pointer
> points to something other that void. Can this really make trouble?
> Anyway, for completeness:
> 
>       2       ../include/getopt.h
>      22       sanei_wire.c
>      22       sanei_scsi.c
>       4       sanei_usb.c
>       2       abaton.c
>       4       agfafocus.c
>       2       apple.c
>       3       artec.c
>       5       as6e.c
>       2       avision.c
>      10       canon-sane.c
>       2       canon630u-common.c
>       1       canon630u.c
>       1       canon_pp-dev.c
>      27       canon_pp.c
>      18       coolscan-scsidef.h
>      64       coolscan.c
>       2       coolscan2.c
>       1       dc25.c
>      10       djpeg.c
>       1       dc210.c
>      17       dmc.c
>       2       epson_scsi.c
>       2       epson.c
>      55       gt68xx_low.c
>       2       gt68xx_gt6801.c
>       2       gt68xx_gt6816.c
>       4       hp-accessor.c
>       4       hp-device.c
>      55       hp-option.c
>       2       hp-scl.c
>       2       hp.c
>       1       leo.c
>       2       matsushita.c
>      11       microtek.c
>      65       microtek2.c
>      33       mustek_pp.c
>       3       mustek_usb_low.c
>       4       mustek_usb_mid.c
>      14       mustek_usb_high.c
>      18       nec.c
>      20       net.c
>       4       pie.c
>       2       plustek.c
>       5       qcam.c
>       1       ricoh-scsi.c
>       2       ricoh.c
>      12       s9036.c
>       1       sceptre.c
>      13       sharp.c
>       4       snapscan-usb.c
>      18       sp15c-scsi.h
>      23       sp15c.c
>       1       sp15c-scsi.h
>       1       sp15c.c
>      33       st400.c
>       8       tamarack.c
>       3       test-picture.c
>      11       test.c
>       1       teco1.c
>       1       teco2.c
>       1       teco3.c
>       3       umax-usb.c
>       1       umax.c
>      14       umax_pp_low.c
>       6       umax_pp.c
>       1       umax1220u.c
>       1       v4l.c
>       1       dll.c
>       2       dll-preload.c
>       9       dll.c
>       3       scanimage.c
>       4       saned.c
> 
> Full log at:
> http://www.meier-geinitz.de/sane/tmp/warnings-2002-11-16-gcc3.0-warn.log
> 
> Sorted by type of warning (gcc-3.0, warnings enabled):
>     230        warning: void format, [foo] type
>     170        warning: unused parameter `[foo]'
>     115        warning: cast discards qualifiers from pointer target type
>      54        warning: pointer targets in passing arg [foo] of `[bar]' 
> differ in signedness
>      37        warning: pointer targets in initialization differ in signedness
>      32        warning: comparison between signed and unsigned
>      19        warning: passing arg [foo] of `[bar]' discards qualifiers from 
> pointer target type
>      11        warning: unused variable `[foo]'
>       5        warning: signed and unsigned type in conditional expression
>       5        warning: function declaration isn't a prototype
>       5        warning: `[foo]' defined but not used
>       4        warning: ISO C forbids use of cast expressions as lvalues
>       3        warning: string length `[foo]' is greater than the length 
> `509' ISO C89 compilers are required to support
>       3        warning: overflow in implicit constant conversion
>       3        warning: implicit declaration of function [foo]
>       3        warning: assignment discards qualifiers from pointer target 
> type
>       2        warning: pointer targets in assignment differ in signedness
>       2        warning: passing arg [foo] of `[bar]' from incompatible 
> pointer type
>       2        warning: missing initializer
>       2        warning: [foo] format, [bar] arg (arg [baz])
>       1        warning: too many arguments for format
>       1        warning: suggest explicit braces to avoid ambiguous `else'
>       1        warning: operation on `p' may be undefined
>       1        warning: assignment from incompatible pointer type
> 
> Bye,
>   Henning
> _______________________________________________
> Sane-devel mailing list
> [email protected]
> http://www.mostang.com/mailman/listinfo/sane-devel
> 


-- 
Peter Kirchgessner
http://www.kirchgessner.net
mailto:[email protected]


Reply via email to