Re: [PATCH 4/4] perf symbols: Increase symbol KSYM_NAME_LEN size

[Ricardo Ribalda Delgado]

Also suck a big binary blob does not help :)

ricardo@newpili:/tmp/fglrx-driver-11-9/arch/x86_64/lib/modules/fglrx/build_mod$
for a in $( objdump -t libfglrx_ip.a | grep text| awk '{print $6}' );
do printf " %s $a\n" $(echo $a|  wc -c) ; done | sort -n  | tee
symbols | tail -n 10
 137 _ZNK11R800AddrLib28ComputeSurfaceInfoMicroTiledEPK32_ADDR_COMPUTE_SURFACE_INFO_INPUTP33_ADDR_COMPUTE_SURFACE_INFO_OUTPUTj13_AddrTileMode
 140 _ZNK11R800AddrLib33DispatchComputeFmaskCoordFromAddrEPK39_ADDR_COMPUTE_FMASK_COORDFROMADDR_INPUTP40_ADDR_COMPUTE_FMASK_COORDFROMADDR_OUTPUT
 141 _ZNK11R600AddrLib30HwlComputeSurfaceAddrFromCoordEPK41_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_INPUTP42_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_OUTPUT
 141 _ZNK11R600AddrLib30HwlComputeSurfaceCoordFromAddrEPK41_ADDR_COMPUTE_SURFACE_COORDFROMADDR_INPUTP42_ADDR_COMPUTE_SURFACE_COORDFROMADDR_OUTPUT
 141 _ZNK11R800AddrLib30HwlComputeSurfaceAddrFromCoordEPK41_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_INPUTP42_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_OUTPUT
 141 _ZNK11R800AddrLib30HwlComputeSurfaceCoordFromAddrEPK41_ADDR_COMPUTE_SURFACE_COORDFROMADDR_INPUTP42_ADDR_COMPUTE_SURFACE_COORDFROMADDR_OUTPUT
 146 _ZNK11R600AddrLib35DispatchComputeSurfaceAddrFromCoordEPK41_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_INPUTP42_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_OUTPUT
 146 _ZNK11R600AddrLib35DispatchComputeSurfaceCoordFromAddrEPK41_ADDR_COMPUTE_SURFACE_COORDFROMADDR_INPUTP42_ADDR_COMPUTE_SURFACE_COORDFROMADDR_OUTPUT
 146 _ZNK11R800AddrLib35DispatchComputeSurfaceAddrFromCoordEPK41_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_INPUTP42_ADDR_COMPUTE_SURFACE_ADDRFROMCOORD_OUTPUT
 146 _ZNK11R800AddrLib35DispatchComputeSurfaceCoordFromAddrEPK41_ADDR_COMPUTE_SURFACE_COORDFROMADDR_INPUTP42_ADDR_COMPUTE_SURFACE_COORDFROMADDR_OUTPUT


On Mon, Oct 24, 2011 at 20:05,  <Valdis.Kletnieks@xxxxxx> wrote:
> On Sun, 23 Oct 2011 17:12:23 -0200, Arnaldo Carvalho de Melo said:
>
>> Fglrx propietary driver has symbol names over 128 chars (:S). This
>> breaks the function kallsyms__parse.
>
> Well, that explains why they haven't upstreamed the driver, it will never
> pass checkpatch's line-lenght check :)
>
> But at some point you really need to just tell the vendors to put down
> the crack pipe.
>
> (What *are* the symbols? C++ mangled function names?)
>



-- 
Ricardo Ribalda
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/