Also this is a quite old question, maybe you or other are interested in a possible way.

First let's create a bed file, containing intervals of covered site in each mpileup file. As this question is 10 month old, I assume you have the output of samtools mpileup where the first column contains the chromosome name and second the position.

$ awk -v OFS="\t" '{print $1,$2-1,$2}' mpileup1 | bedtools merge > mpileup_file1.bed
$ awk -v OFS="\t" '{print $1,$2-1,$2}' mpileup2 | bedtools merge > mpileup_file2.bed

Now you can intersect these to files, to find common regions:

$ bedtools intersect -a mpileup_file1.bed -b mpileup_file2.bed > mpileup_intersect.bed

To extend these common intervals you can use bedtools slop, which would require a file genome file. Or you use a quick-and-dirty awk solution:

$ awk -v OFS="\t" '{print $1, $2-100, $3+100}' mpileup_intersect.bed > mpileup_padded.bed

The next step is to compress your mpileup files using bgzip and index them with tabix.

$ bgzip -c mpileup1 > mpileup1.gz
$ bgzip -c mpileup2 > mpileup2.gz

$ tabix -s 1 -b 2 -e 2 mpileup1.gz
$ tabix -s 1 -b 2 -e 2 mpileup2.gz

Now we are able, to quickly receive the information at the positions given in our mpileup_padded.bed.

$ tabix -R mpileup_padded.bed mpileup1.gz
$ tabix -R mpileup_padded.bed mpileup2.gz

Use the output how ever you like.

fin swimmer