version 1.269, 2008/04/16 18:50:38 |
version 1.270, 2008/04/25 00:50:55 |
|
|
exposed more often by the 64-bit big endian nature of UltraSPARC. |
exposed more often by the 64-bit big endian nature of UltraSPARC. |
|
|
<p> |
<p> |
Sun refused access to the neccessary documentation for the (very bizzare) |
Sun refused access to the necessary documentation for the (very bizarre) |
host bridge and caches in the UltraSPARC III machines, so a few years were |
host bridge and caches in the UltraSPARC III machines, so a few years were |
lost before some reverse engineering figured out the changes in these |
lost before some reverse engineering figured out the changes in these |
machines (and the undocumented processor bugs, too). In 2007, Sun |
machines (and the undocumented processor bugs, too). In 2007, Sun |
|
|
<h3><font color="#0000e0"><strong>Hardware documentation:</strong></font></h3> |
<h3><font color="#0000e0"><strong>Hardware documentation:</strong></font></h3> |
|
|
Previous to 2007, it was extremely difficult or impossible to get much of |
Previous to 2007, it was extremely difficult or impossible to get much of |
the neccessary hardware documentation for these machines. Members of our |
the necessary hardware documentation for these machines. Members of our |
team (in particular David Gwynne) were instrumental in pressuring and |
team (in particular David Gwynne) were instrumental in pressuring and |
convincing Sun to go back into their archives and dig up the data sheets |
convincing Sun to go back into their archives and dig up the data sheets |
that they had written in-house, describing the operation of their chips to |
that they had written in-house, describing the operation of their chips to |