Re: [Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
On Thu, 2015-01-22 at 15:57 +, Ian Campbell wrote: On Thu, 2015-01-22 at 15:52 +, Ian Campbell wrote: Pushing the ~1 to the output will force a test run, once that is done I will push the osstest patches. Actually using the result will require me to push this along with the main series. I've have just pushed these to pretest as discussed last week: $ git log --oneline bb2240e..2e35f88 2e35f88 make-flight: reorganize scheduling related test jobs 25ee621 Switch ARM testing back to linux-arm-xen branch. 339ad81 Debian: create boot.scr.nat when installing Xen on u-boot. fcf00b2 TestSupport: replace power_cycle_timeout with power_cycle_sleep 70b19e1 Debian: Create boot.scr with a suffix and copy to boot.scr e449fed ts-kernel-build: Adjust kernel .config to work on the arndale boards. ee5e442 Osstest/Debian: Add 0x prefix to $filesize 0e24cb2 Osstest/Debian: Add clk_ignore_unused to default command line 653d0af Osstest/Debian: Remove hardcoded addresses from boot.scr 2d08c4c Osstest/Debian: support adding a rootdelay property to bootargs 64c57cb Osstest/Debian.pm: Refactor common kernel cmdline setup. 45e3b3c Osstest/Debian: Refactor setting of dom0 kernel command line. 070db63 Osstest/Debian: Support for loading an FDT from u-boot script 781c575 Osstest/Debian: Install dtbs into target filesystem in /boot/dtbs 2d92112 Osstest/PDU: Add eth008.pm method to control the ARM rack PDU 056840c make-flight: Run a basic test on each arm platform That is: This series, the one extra patch to use linux-arm-xen for kernels and Dario's scheduler rejigging. I also ran: OSSTEST_CONFIG=production-config ./mg-hosts setflags marilith-n\* -- platform-midway So that 056840c does something. Ian. ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Re: [Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
On Fri, 2015-01-23 at 12:03 +, Stefano Stabellini wrote: You are right, I was mistaken. However keep in mind that only the ubuntu stable branch linux-3.16.y-queue has the following important backport: commit 342443e998bd2dd6406d46c5f7e62540a6d4a170 Author: Stefano Stabellini stefano.stabell...@eu.citrix.com Date: Fri Nov 21 11:07:39 2014 + xen/arm/arm64: introduce xen_arch_need_swiotlb commit a4dba130891271084344c12537731542ec77cb85 upstream. Introduce an arch specific function to find out whether a particular dma mapping operation needs to bounce on the swiotlb buffer. the main linux-3.16.y branch (currently tagged v3.16.7-ckt4) doesn't have it yet. Without it, DMA is not safe with non-coherent devices. Hrm, how unfortunate. We expect that patch to land in ckt5 I suppose? I think we can either wait for ckt5 or you could apply it to your branch, which means future versions would need merging not fast-forwarding. Not horrendous I suppose. ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Re: [Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
On Fri, 23 Jan 2015, Ian Campbell wrote: On Thu, 2015-01-22 at 18:54 +, Stefano Stabellini wrote: On Thu, 22 Jan 2015, Ian Campbell wrote: On Thu, 2015-01-22 at 10:56 +, Ian Campbell wrote: I've managed to make a habit of forgetting to mention one important point (including hitting send on this bit in reply to v4!) Both the cubietruck and the arndale are (for different reasons) not supported by the 3.14 kernel which we currently use as a baseline for osstest testing. I'm not sure if another LTS kernel is due from upstream soon or not, if not then I think we will need to resurrect the linux-arm-xen branch and use it for all arm platforms (including midway, since I don't think we want to take on the complexity of differing per platform). So far I've been testing using v3.16.7, which is the last 3.16 longterm release, however canonical are carrying it on at http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.16.y (up to 3.16.7-ckt4 at the moment). This is the same base as Debian will be shipping in Jessie. So that is one option. Stefano, do you have another preference? We discussed this IRL and are going to resurrect the linux-arm-xen branch following the Canonical stable v3.16 branch. Stefano, Please could you pull from git://kernel.ubuntu.com/ubuntu/linux.git and the force push:: Actually I think we should follow git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git linux-3.16.y There are no more updates to this, since it is not a kernel.org LTS kernel. The canonical kernel team (ckt in the tag name) are doing ongoing stable support for these kernels (which is also being consumed by Debian). The commit pointed to by tag v3.16.7-ckt4 (i.e. v3.16.7-ckt4~0) to the I was talking about v3.16.7 This doesn't have e.g. the swiotlb/dma fixes which you have had backported, those are in the ckt kernels only. You are right, I was mistaken. However keep in mind that only the ubuntu stable branch linux-3.16.y-queue has the following important backport: commit 342443e998bd2dd6406d46c5f7e62540a6d4a170 Author: Stefano Stabellini stefano.stabell...@eu.citrix.com Date: Fri Nov 21 11:07:39 2014 + xen/arm/arm64: introduce xen_arch_need_swiotlb commit a4dba130891271084344c12537731542ec77cb85 upstream. Introduce an arch specific function to find out whether a particular dma mapping operation needs to bounce on the swiotlb buffer. the main linux-3.16.y branch (currently tagged v3.16.7-ckt4) doesn't have it yet. Without it, DMA is not safe with non-coherent devices. ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Re: [Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
On Thu, 2015-01-22 at 18:54 +, Stefano Stabellini wrote: On Thu, 22 Jan 2015, Ian Campbell wrote: On Thu, 2015-01-22 at 10:56 +, Ian Campbell wrote: I've managed to make a habit of forgetting to mention one important point (including hitting send on this bit in reply to v4!) Both the cubietruck and the arndale are (for different reasons) not supported by the 3.14 kernel which we currently use as a baseline for osstest testing. I'm not sure if another LTS kernel is due from upstream soon or not, if not then I think we will need to resurrect the linux-arm-xen branch and use it for all arm platforms (including midway, since I don't think we want to take on the complexity of differing per platform). So far I've been testing using v3.16.7, which is the last 3.16 longterm release, however canonical are carrying it on at http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.16.y (up to 3.16.7-ckt4 at the moment). This is the same base as Debian will be shipping in Jessie. So that is one option. Stefano, do you have another preference? We discussed this IRL and are going to resurrect the linux-arm-xen branch following the Canonical stable v3.16 branch. Stefano, Please could you pull from git://kernel.ubuntu.com/ubuntu/linux.git and the force push:: Actually I think we should follow git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git linux-3.16.y There are no more updates to this, since it is not a kernel.org LTS kernel. The canonical kernel team (ckt in the tag name) are doing ongoing stable support for these kernels (which is also being consumed by Debian). The commit pointed to by tag v3.16.7-ckt4 (i.e. v3.16.7-ckt4~0) to the I was talking about v3.16.7 This doesn't have e.g. the swiotlb/dma fixes which you have had backported, those are in the ckt kernels only. Ian. ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
[Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
I'm preparing a bunch of each of these boards to be racked for inclusion in osstest. This series adds the necessary osstest support: * Support for platforms which require us to supply a Device Tree * Removing various hardcoded midway assumptions * Some new features needed to workaround h/w quirks (of arndale in particular) * Support for the power relay switch board which these are attached to * Some new convenience features The series is completely acked now, reposting just because some of the acks were given preemptively to give a chance for objections etc. I've managed to make a habit of forgetting to mention one important point (including hitting send on this bit in reply to v4!) Both the cubietruck and the arndale are (for different reasons) not supported by the 3.14 kernel which we currently use as a baseline for osstest testing. I'm not sure if another LTS kernel is due from upstream soon or not, if not then I think we will need to resurrect the linux-arm-xen branch and use it for all arm platforms (including midway, since I don't think we want to take on the complexity of differing per platform). So far I've been testing using v3.16.7, which is the last 3.16 longterm release, however canonical are carrying it on at http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.16.y (up to 3.16.7-ckt4 at the moment). This is the same base as Debian will be shipping in Jessie. So that is one option. Stefano, do you have another preference? WRT deploying we need to figure out the sequence in which we push these patches, reenable the linux-arm-xen branch and commissioning the new boards. Ian. PlatformsArmhf midway cubietruck arndale HostProp_arndale-metrocentre_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 1 HostProp_arndale-westfield_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 2 HostProp_arndale-lakeside_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 3 HostProp_arndale-bluewater_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 4 HostProp_cubietruck-braque_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 5 HostProp_cubietruck-picaso_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 6 HostProp_cubietruck-metzinger_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 7 HostProp_cubietruck-gleizes_PowerMethod eth008 arm-pdu-01.cam.xci-test.com admin password 8 HostGroupProp_arndale_LinuxSerialConsole ttySAC2 HostGroupProp_arndale_Build_Make_Flags -j4 HostGroupProp_arndale_XenSerialConsole dtuart HostGroupProp_arndale_XenDTUARTPath /serial@12C2 HostGroupProp_arndale_Interface_Force eth0 HostGroupProp_arndale_ExtraInitramfsModules clk-s2mps11 s5m8767 i2c-s3c2410 phy-exynos5250-sata HostGroupProp_arndale_Rootdelay 3 HostGroupProp_arndale_UBootSetXenAddrR 0x4100 HostGroupProp_arndale_Serial sympathy osstser1 HostGroupProp_arndale_PowerCycleTime 15 HostGroupProp_arndale_TftpScope xenclient HostGroupFlags_arndale suite-wheezy,platform-arndale,need-kernel-deb-armmp,no-di-kernel,force-mac-address,need-uboot-bootscr HostGroup_arndale-metrocentre arndale HostProp_arndale-metrocentre_Fqdn arndale-metrocentre.cam.xci-test.com HostProp_arndale-metrocentre_Ether 9e:04:00:59:64:5a HostGroup_arndale-westfield arndale HostProp_arndale-westfield_Fqdn arndale-westfield.cam.xci-test.com HostProp_arndale-westfield_Ether a6:46:13:77:e8:2f HostGroup_arndale-lakeside arndale HostProp_arndale-lakeside_Fqdn arndale-lakeside.cam.xci-test.com HostProp_arndale-lakeside_Ether e2:e7:75:3f:df:4c HostGroup_arndale-bluewater arndale HostProp_arndale-bluewater_Fqdn arndale-bluewater.cam.xci-test.com HostProp_arndale-bluewater_Ether f2:dc:22:d7:e9:e9 HostGroupProp_cubietruck_LinuxSerialConsole ttyS0 HostGroupProp_cubietruck_Build_Make_Flags -j4 HostGroupProp_cubietruck_XenSerialConsole dtuart HostGroupProp_cubietruck_XenDTUARTPath /soc@01c0/serial@01c28000 HostGroupProp_cubietruck_UBootSetXenAddrR 0x4100 HostGroupProp_cubietruck_Serial sympathy arm-serial-01 HostGroupProp_cubietruck_TftpScope xenclient HostGroupFlags_cubietruck suite-wheezy,platform-cubietruck,need-kernel-deb-armmp,no-di-kernel,need-uboot-bootscr HostGroup_cubietruck-braque cubietruck HostProp_cubietruck-braque_Fqdn cubietruck-braque.cam.xci-test.com HostGroup_cubietruck-picaso cubietruck HostProp_cubietruck-picaso_Fqdn cubietruck-picaso.cam.xci-test.com HostGroup_cubietruck-metzinger cubietruck HostProp_cubietruck-metzinger cubietruck-metzinger.cam.xci-test.com HostGroup_cubietruck-gleizes cubietruck HostProp_cubietruck-gleizes_Fqdn cubietruck-gleizes.cam.xci-test.com ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Re: [Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
On Thu, 22 Jan 2015, Ian Campbell wrote: On Thu, 2015-01-22 at 10:56 +, Ian Campbell wrote: I've managed to make a habit of forgetting to mention one important point (including hitting send on this bit in reply to v4!) Both the cubietruck and the arndale are (for different reasons) not supported by the 3.14 kernel which we currently use as a baseline for osstest testing. I'm not sure if another LTS kernel is due from upstream soon or not, if not then I think we will need to resurrect the linux-arm-xen branch and use it for all arm platforms (including midway, since I don't think we want to take on the complexity of differing per platform). So far I've been testing using v3.16.7, which is the last 3.16 longterm release, however canonical are carrying it on at http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.16.y (up to 3.16.7-ckt4 at the moment). This is the same base as Debian will be shipping in Jessie. So that is one option. Stefano, do you have another preference? We discussed this IRL and are going to resurrect the linux-arm-xen branch following the Canonical stable v3.16 branch. Stefano, Please could you pull from git://kernel.ubuntu.com/ubuntu/linux.git and the force push:: Actually I think we should follow git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git linux-3.16.y The commit pointed to by tag v3.16.7-ckt4 (i.e. v3.16.7-ckt4~0) to the I was talking about v3.16.7 input tree#branch at: git://git.kernel.org/pub/scm/linux/kernel/git/sstabellini/xen.git#linux-arm-xen and v3.16.7-ckt4~1 (i.e. the commit before the tag) to the output tree#branch at: git://xenbits.xen.org/linux-pvops.git#tested/linux-arm-xen Pushing the ~1 to the output will force a test run, once that is done I will push the osstest patches. Ian. ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Re: [Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
On Thu, 2015-01-22 at 10:56 +, Ian Campbell wrote: I've managed to make a habit of forgetting to mention one important point (including hitting send on this bit in reply to v4!) Both the cubietruck and the arndale are (for different reasons) not supported by the 3.14 kernel which we currently use as a baseline for osstest testing. I'm not sure if another LTS kernel is due from upstream soon or not, if not then I think we will need to resurrect the linux-arm-xen branch and use it for all arm platforms (including midway, since I don't think we want to take on the complexity of differing per platform). So far I've been testing using v3.16.7, which is the last 3.16 longterm release, however canonical are carrying it on at http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.16.y (up to 3.16.7-ckt4 at the moment). This is the same base as Debian will be shipping in Jessie. So that is one option. Stefano, do you have another preference? We discussed this IRL and are going to resurrect the linux-arm-xen branch following the Canonical stable v3.16 branch. Stefano, Please could you pull from git://kernel.ubuntu.com/ubuntu/linux.git and the force push:: The commit pointed to by tag v3.16.7-ckt4 (i.e. v3.16.7-ckt4~0) to the input tree#branch at: git://git.kernel.org/pub/scm/linux/kernel/git/sstabellini/xen.git#linux-arm-xen and v3.16.7-ckt4~1 (i.e. the commit before the tag) to the output tree#branch at: git://xenbits.xen.org/linux-pvops.git#tested/linux-arm-xen Pushing the ~1 to the output will force a test run, once that is done I will push the osstest patches. Ian. ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Re: [Xen-devel] [PATCH OSSTEST v5 00/13] support for ARM32 arndale and cubietruck platforms
On Thu, 2015-01-22 at 15:52 +, Ian Campbell wrote: Pushing the ~1 to the output will force a test run, once that is done I will push the osstest patches. Actually using the result will require me to push this along with the main series. 8--- From ba72d891c2d97991a4b00d8958883d17e4bc16d1 Mon Sep 17 00:00:00 2001 From: Ian Campbell ian.campb...@citrix.com Date: Thu, 22 Jan 2015 15:53:35 + Subject: [PATCH] Switch ARM testing back to linux-arm-xen branch. The cubietruck and arndale boards need a newer kernel than our current baseline of 3.14. Therefore switch all ARM testing to the linux-arm-kernel branch which has been moved up to test v3.16. This effectively reverts commit 38b2fe2020883 Switch armhf to use common default Linux. Signed-off-by: Ian Campbell ian.campb...@citrix.com --- mfi-common | 11 ++- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/mfi-common b/mfi-common index e167606..7ccff40 100644 --- a/mfi-common +++ b/mfi-common @@ -73,6 +73,10 @@ create_build_jobs () { xen-4.1-testing) continue;; xen-4.2-testing) continue;; esac + pvops_kernel= +tree_linux=$TREE_LINUX_ARM +revision_linux=${REVISION_LINUX_ARM:-${DEFAULT_REVISION_LINUX_ARM}} + pvops_kconfig_overrides= kconfig_override_y=CONFIG_EXT4_FS @@ -81,15 +85,12 @@ create_build_jobs () { case $branch in linux-arm-xen) continue;; esac - ;; -esac - -if [ x$pvops_kernel = x ] ; then pvops_kernel= tree_linux=$TREE_LINUX revision_linux=${REVISION_LINUX:-${DEFAULT_REVISION_LINUX}} -fi + ;; +esac case $arch in *) suite=$defsuite;; -- 2.1.4 ___ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel