On 12.10.2022 08:42, Henry Wang wrote: > Hi Jan, > >> -----Original Message----- >> From: Jan Beulich <[email protected]> >> Subject: Re: [xen-4.15-testing test] 173498: regressions - FAIL >> >> On 12.10.2022 04:42, Henry Wang wrote: >>>> -----Original Message----- >>>> Subject: [xen-4.15-testing test] 173498: regressions - FAIL >>>> >>>> flight 173498 xen-4.15-testing real [real] >>>> http://logs.test-lab.xenproject.org/osstest/logs/173498/ >>>> >>>> Regressions :-( >>> >>> I think these regressions are from the backporting happened yesterday, >>> see below... >>> >>>> >>>> Tests which did not succeed and are blocking, >>>> including tests which could not be run: >>>> build-arm64 6 xen-build fail REGR. vs. >>>> 172547 >>>> build-arm64-xsm 6 xen-build fail REGR. vs. >>>> 172547 >>>> build-amd64 6 xen-build fail REGR. vs. >>>> 172547 >>>> build-armhf 6 xen-build fail REGR. vs. >>>> 172547 >>> >>> ...The arm/arm64 regression is from the backporting of commit: >>> xen/arm, libxl: Implement XEN_DOMCTL_shadow_op for Arm >>> >>> The issue is: >>> In 4.16, commit >>> 2107cc76db3a libxc: split xc_logdirty_control() from xc_shadow_control() >>> changes the prototype of xc_shadow_control(), and hence the calling of >>> xc_shadow_control() in 4.13, 4.14 and 4.15 does not match the calling of >>> xc_shadow_control() in 4.16 and after. >> >> So did I commit a wrong patch, or were the patches for the older branches >> wrong already in xsa.git? > > I think the latter. For this commit I think the patch for master branch was > picked > for the 4.13-4.15 backporting, without noticing the xc_shadow_control() > prototype change in 4.16. > > I can send a fix for this. If we don't want to backport the > "libxc: split xc_logdirty_control() from xc_shadow_control()"
I don't think we should backport that extra change, but Anthony would have the final say. Jan
