Current-Users archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: Yet another build problem for evbppc64
On Sat, 24 Nov 2012, Paul Goyette wrote:
With up-to-date sources...
--- dependall-../external/bsd/mdocml/lib ---
/var/tmp//cc3rajhb.s: Assembler messages:
/var/tmp//cc3rajhb.s:90: Error: symbol `.data' is already defined
*** [tbl_data.o] Error code 1
nbmake: stopped in /build/netbsd-local/src/external/bsd/mdocml/lib/libmandoc
Looking at the compiler output, I find that we have (starting at line
83)
.section ".opd","aw"
.align 3
data:
.quad .data,.TOC.@tocbase,0
.previous
.size data,24
.type .data,@function
.data:
mflr 0
std 27,-40(1)
std 28,-32(1)
std 29,-24(1)
std 30,-16(1)
std 31,-8(1)
...
.L49:
ld 3,.LC3@toc(2)
ld 5,.LC4@toc(2)
li 4,79
ld 6,.LC6@toc(2)
bl .__assert13
nop
.long 0
.byte 0,0,0,1,128,8,0,0
.size .data,.-.data
.section ".toc","aw"
.LC18:
.tc .LC17[TC],.LC17
.section ".text"
.align 2
.p2align 4,,15
.globl tbl_cdata
.section ".opd","aw"
...
The definition of .data is at line 90.
Unless there is some internal definition of the symbol, I don't see how
this could be "already defined".
However, since the function data() within this source file is defined as
static, is there any reason why it could not be renamed to something
else just to avoid the internal name-space conflict?
The attached diff renames the function to xdata() and this allows the
build to proceed. Does anyone object to committing it?
-------------------------------------------------------------------------
| Paul Goyette | PGP Key fingerprint: | E-mail addresses: |
| Customer Service | FA29 0E3B 35AF E8AE 6651 | paul at whooppee.com |
| Network Engineer | 0786 F758 55DE 53BA 7731 | pgoyette at juniper.net |
| Kernel Developer | | pgoyette at netbsd.org |
-------------------------------------------------------------------------
Home |
Main Index |
Thread Index |
Old Index