Understanding defect histories of CGTs

alex kavcak

Looking at the defect histories for the compiler generation tools. Is it
fair to assume that any problem fixed in one particular version is present
in every previous versions?

Specifically my interest is for the C6000 7.3.x releases where x = 0, 1, .. 14

How about moving from 7.3 to 7.4? Does 7.3 has all the bugs fixed in 7.4 versions?

Thanks

over 11 years ago

0 Archaeologist over 11 years ago

TI__Guru* 84225 points

No. Please see http://processors.wiki.ti.com/index.php/Compiler_Version_Numbers_and_What_They_Mean

It is possible for a bug to be introduced (or exposed) in any release, but it's very unlikely from one patch release to the next. That is, 7.3.14 is only supposed to contain bug fixes from 7.3.13, and is thus very unlikely to introduce a new bug. 7.4.0 introduces new features, which inevitably adds some risk.

To know whether a bug reported as fixed in 7.4.0 exists in 7.3.x, you'll have to look in SDOWP to see which versions the defect was reported for.

0 alex kavcak over 11 years ago in reply to Archaeologist

Prodigy 90 points

Where can I find the "SDOWP"?

My bugfix experience tells me that a 'bug' was present before it was noticed the first time. The question than is how long was it present. And I understand well that new features can introduce issues, but probably not every new bug is a product of new feature.

0 Archaeologist over 11 years ago in reply to alex kavcak

TI__Guru* 84225 points

You are correct that bugs are often present before they were noticed for the first time. There's no way to definitively determine in which versions a bug is present without considering each bug individually.

For SDOWP, please see http://processors.wiki.ti.com/index.php/Compiler_Defects_Database

Code Composer Studio™︎

Code Composer Studio forum

Understanding defect histories of CGTs