> As it has been a while since anyone has posted 4.2.0 testsuite for cygwin, > your comparison against past posted results for 4.2 (and more recently, 4.3.0) > is as good as mine. Your results look reasonable. > Once in a very long time (like 2 years or so) someone familiar with the particular > tests may ask for more specific detail. Tim, Thanks for the reply. I picked 4.2.0 because according to gcc.gnu.org it is the "current release". Although, since I have to operate in a cygwin environment I am having my doubts. Being new to this territory I did some superficial comparisons against the 4 other results for 4.2.0/i686/cygwin that I could find since 1-Jan-07. Since there are so few recent ones I posted mine as well. Attached is a table of these 5 sets. It is superficial in that similar to looking at MLS listings for houses all I can see is the # of bedrooms, bathrooms, floorspace, etc. I have little insight into what is inside these rooms. It appears that my results compare reasonably to Rob1weld's and Christian Joensson's May 24th results. What is baffling is that RW's libstdc++ testing produced 50 more tests than mine and his gcc tests produced 738 more tests! Where did they go? Why wouldn't they show up as at least Untested? How do I find/apply Dave Korn's patch to stdio.h / newlib? Is it incorporated into any Cygwin contributed snapshots, or is it a pure manual source edit? Best regards, Lee.
Attachment:
GCC_4.2.0_Cygwin_comparisons.pdf
Description: Adobe PDF document
