[Patch]: ld - add --print-memory-usage option

Nicholas Clifton nickc@redhat.com
Wed Jun 10 15:19:00 GMT 2015


Hi Tristan,

> In fact I added three testcases :-)

Thanks very much.  There is just one small problem.  Some regressions:

  alpha-linuxecoff ...  LD REGRESSION: print-memory-usage-1
  alpha-unknown-osf4.0 ... LD REGRESSION: print-memory-usage-1
  arm-aout ... LD REGRESSION: print-memory-usage-1
  i386-pc-go32 ... LD REGRESSION: print-memory-usage-1
  i586-aout ... LD REGRESSION: print-memory-usage-1
  mips-elf ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  mips-sgi-irix6 ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  mips64-linux ... LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  mips64vrel-elf ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  mipsel-linux-gnu ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  mipsisa32el-linux ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  ns32k-netbsd ...  LD REGRESSION: print-memory-usage-1
  pdp11-dec-aout ... LD REGRESSION: print-memory-usage-3
  powerpc-ibm-aix5.2.0 ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  powerpc64-linux ... LD REGRESSION: print-memory-usage-1
  rs6000-aix4.3.3 ... LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  sh-pe ...  LD REGRESSION: print-memory-usage-1, print-memory-usage-2, 
print-memory-usage-3
  spu-elf ... LD REGRESSION: print-memory-usage-1, print-memory-usage-3
  tic30-unknown-aout ... LD REGRESSION: print-memory-usage-1, 
print-memory-usage-2, print-memory-usage-3
  tic4x-coff ... LD REGRESSION: print-memory-usage-1
  tic54x-coff ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  tx39-elf ...  LD REGRESSION: print-memory-usage-1, 
print-memory-usage-3
  x86_64-pc-cygwin ...  LD REGRESSION: print-memory-usage-1


I can provide you with the full logs if you like.  A quick look however 
appears to suggest that some targets are increasing the sizes of the 
memory regions, possibly for alignment purposes, and that causes the 
regexps to fail.

Cheers
   Nick



More information about the Binutils mailing list