Conversation
|
REQUEST FOR PRODUCTION RELEASES: This will add The following labels are available |
|
needs something like #15162 |
Signed-off-by: Felix Schlepper <felix.schlepper@cern.ch>
Signed-off-by: Felix Schlepper <felix.schlepper@cern.ch>
|
Error while checking build/O2/fullCI_slc9 for a3b972a at 2026-03-25 22:10: Full log here. |
Removed math.h and added cmath for better compatibility.
|
Error while checking build/O2/fullCI_slc9 for 7c07166 at 2026-03-26 18:16: Full log here. |
|
@shahor02 can we please merge this, the failing tests should be unrelated. |
shahor02
left a comment
There was a problem hiding this comment.
Merging, one insignifican comment and one question below (will have more later)
| skew << 0, -posLoc[2], posLoc[1], posLoc[2], 0, -posLoc[0], -posLoc[1], posLoc[0], 0; | ||
| jac.topLeftCorner<3, 3>() = rotL2T; | ||
| jac.topRightCorner<3, 3>() = -rotL2T * skew; | ||
| jac.bottomRightCorner<3, 3>() = rotL2T; |
There was a problem hiding this comment.
Could you comment on the logic behind this?
This removes entirely the previous misalignment code (that's why the digitiser is touched), adds the possibility to study different mis-alignment modes and its effect on the residuals and implements a more or less finished prototype to interface with GBL&MillePede to find back the injected mis-alignment.