ni / linux

Linux kernel source for NI Linux Real-Time
Other
80 stars 77 forks source link

Merge tag 'v5.15.129-rt67' into nilrt/master/5.15 #134

Closed mike-petersen-ni closed 1 year ago

mike-petersen-ni commented 1 year ago

Merge in preparation for contingency due to performance regressions in NI drivers. AB#2519871

Conflicts

There were two conflicts.

drivers/usb/dwc3/core.h

line 1088@ours, 1085@theirs, 1085@resolved

++<<<<<<< ours
 + * @has_dsm_for_softreset: set if we want to use BIOS to do core soft reset
++||||||| base
++=======
+  * @suspended: set to track suspend event due to U3/L2.
++>>>>>>> theirs

Resolved by concatenating, matching them to the order used in the struct. This appears to match our nilrt/master/6.1 branch.

drivers/usb/dwc3/dwc3-pci.c

line 222@ours, 221@theirs, 222@resolved

++<<<<<<< ours
++||||||| base
++           * A lot of BYT devices lack ACPI resource entries for
++           * the GPIOs, add a fallback mapping to the reference
++           * design GPIOs which all boards seem to use.
++           */
++          gpiod_add_lookup_table(&platform_bytcr_gpios);
++
++          /*
++=======
+            * A lot of BYT devices lack ACPI resource entries for
+            * the GPIOs. If the ACPI entry for the GPIO controller
+            * is present add a fallback mapping to the reference
+            * design GPIOs which all boards seem to use.
+            */
+           if (acpi_dev_present("INT33FC", NULL, -1))
+               gpiod_add_lookup_table(&platform_bytcr_gpios);
+ 
+           /*
++>>>>>>> theirs

Resolved by reverting 4451ed9ec933 before merging, as theirs appears to be an upstreamed patch from us (6f126e026307). This appears to match the nilrt/master/6.1 branch.

Testing

mike-petersen-ni commented 1 year ago

patch v2

gratian commented 1 year ago

I was trying to test this with CONFIG_NO_HZ_FULL=y and for some reason it hangs on a cRIO-9030 pretty early in the boot. The kernel then reboots after 30 seconds or so probably due to the PANIC_ON_OOPS timeout. Strangely enough it happens only on the second boot into the new kernel (the first boot seems a bit slow but completes).

Still trying to get more information but wanted to give you a heads-up.