Bug#1028343: u-boot-qemu 2023.01~rc4+dfsg-2: riscv64: FDT creation failed
Control: tags -1 upstream X-Debbugs-CC: debian-ri...@lists.debian.org On Wed, Jan 11, 2023 at 09:21:42AM -0800 Vagrant Cascadian wrote: > Definitely would be good to mention to upstream. Please Cc me if you do! I've sent the bugreport upstream: https://lists.denx.de/pipermail/u-boot/2023-January/504466.html Regards, Karsten -- Hiermit widerspreche ich ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#1028343: u-boot-qemu 2023.01~rc4+dfsg-2: riscv64: FDT creation failed
X-Debbugs-CC: debian-ri...@lists.debian.org On Tue, Jan 10, 2023 at 09:22:02PM +0100 Karsten Merker wrote: > I've also taken a look at the u-boot changelogs and there have > been quite a few changes concerning u-boot's handling of > device-trees between the working and the non-working versions. > Unfortunately I'm not familiar enough with the inner workings of > u-boot to understand the implications of several of these > changes. Hello, I've tried narrowing down the source of the issue by using git bisect on the u-boot tree and that has resulted in the following commit as the potential culprit: commit a56f663f07073713042bb0fd08053aeb667e717b Author: Simon Glass Date: Thu Oct 20 18:23:14 2022 -0600 vbe: Add info about the VBE device to the fwupd node At present we put the driver in the /chosen node in U-Boot. This is a bit strange, since U-Boot doesn't normally use that node itself. It is better to put it under the bootstd node. To make this work we need to copy create the node under /chosen when fixing up the device tree. Copy over all the properties so that fwupd knows what to do. Update the sandbox device tree accordingly. Signed-off-by: Simon Glass I'm not sure whether this is the actual culprit or just an operation that happens to expose a problem elsewhere, though. I'm inclined to forward the bug report to u-boot upstream unless somebody has another idea how to get this narrowed down further. Regards, Karsten -- Hiermit widerspreche ich ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#1028343: u-boot-qemu 2023.01~rc4+dfsg-2: riscv64: FDT creation failed
Control: found -1 2023.01+dfsg-1 X-Debbugs-CC: debian-ri...@lists.debian.org On Tue, Jan 10, 2023 at 09:23:04AM -0800 Vagrant Cascadian wrote: > On 2023-01-10, Cheng Li wrote: [...] > > Moving Image from 0x8400 to 0x8020, end=815d8000 > > ## Flattened Device Tree blob at ff733ef0 > > Booting using the fdt blob at 0xff733ef0 > > Working FDT set to ff733ef0 > > Using Device Tree in place at ff733ef0, end ff738dd2 > > Working FDT set to ff733ef0 > > ERROR: fdt fixup event failed: -22 > > - must RESET the board to recover. > > > > FDT creation failed! hanging...### ERROR ### Please RESET the board ### > > I wonder if this is relevent? > > docs/firmware: Update FW_JUMP documentation > > https://github.com/riscv-software-src/opensbi/commit/7105c189f67028ef73720d7e9816c800ab53dda5 > > Basically changing the address offsets to avoid clobbering one another... Hello, some clobbering could indeed be the source of the problem, though I currently fail to see why that would happen. When running the objdump dance on the working u-boot 2022.10, this gives me the following address blocks: 0x802001a4 0x80200e60 0x80259bbc 0x80275a38 0x802769d1 0x802770c8 0x802778b8 0x80283d60 0x80283e70 0x80284628 0x80287f88 0x80288138 0x8029c2b0 0x8029d9a8 0x802a8008 On the non-working u-boot 2023.01 that is in unstable since today this gives me 0x802001a4 0x80200e70 0x8025a604 0x802762f4 0x802772e8 0x802779ec 0x802781f4 0x802847a8 0x802848b8 0x80285098 0x80288a08 0x80288bb8 0x8029cf40 0x8029e6b0 0x802a8d08 i.e. the used area is a little bit larger in the non-working case, but still way below one MB. On the other hand OpenSBI's ./platform/generic/config.mk contains FW_JUMP_FDT_ADDR=$(shell printf "0x%X" $$(($(FW_TEXT_START) + 0x220))) i.e. AIUI it reserves an area of 0x220 bytes (34MB) for it's payload, so there should be plenty of space and no clobbering should occur. I've also taken a look at the u-boot changelogs and there have been quite a few changes concerning u-boot's handling of device-trees between the working and the non-working versions. Unfortunately I'm not familiar enough with the inner workings of u-boot to understand the implications of several of these changes. Regards, Karsten -- Hiermit widerspreche ich ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#1028343: u-boot-qemu 2023.01~rc4+dfsg-2: riscv64: FDT creation failed
Source: u-boot Version: 2023.01~rc4+dfsg-2 Severity: important User: debian-ri...@lists.debian.org Usertags: riscv64 X-Debbugs-CC: debian-ri...@lists.debian.org Hello, when booting a riscv64 VM in qemu 1:7.0+dfsg-1 with OpenSBI 1.1-2 and u-boot-qemu 2023.01~rc4+dfsg-2 with the following commandline qemu-system-riscv64 -smp 4 -nographic -machine virt -m 8G -bios /usr/lib/riscv64-linux-gnu/opensbi/generic/fw_jump.elf -kernel /usr/lib/u-boot/qemu-riscv64_smode/u-boot.bin -object rng-random,filename=/dev/urandom,id=rng0 -device virtio-rng-device,rng=rng0 -append "console=ttyS0 rw root=/dev/vda1 scsi_mod.use_blk_mq=0" -device virtio-blk-device,drive=hd0 -drive file=riscv64.img,format=raw,id=hd0 -device virtio-net-device,netdev=usernet -netdev user,id=usernet,hostfwd=tcp::2-:22 the following error occurs: -8<--8<--8<--8<--8<- OpenSBI v1.1 _ _ / __ \ / | _ \_ _| | | | |_ __ ___ _ __ | (___ | |_) || | | | | | '_ \ / _ \ '_ \ \___ \| _ < | | | |__| | |_) | __/ | | |) | |_) || |_ \/| .__/ \___|_| |_|_/|/_| | | |_| Platform Name : riscv-virtio,qemu Platform Features : medeleg Platform HART Count : 4 Platform IPI Device : aclint-mswi Platform Timer Device : aclint-mtimer @ 1000Hz Platform Console Device : uart8250 Platform HSM Device : --- Platform Reboot Device: sifive_test Platform Shutdown Device : sifive_test Firmware Base : 0x8000 Firmware Size : 312 KB Runtime SBI Version : 1.0 Domain0 Name : root Domain0 Boot HART : 2 Domain0 HARTs : 0*,1*,2*,3* Domain0 Region00 : 0x0200-0x0200 (I) Domain0 Region01 : 0x8000-0x8007 () Domain0 Region02 : 0x-0x (R,W,X) Domain0 Next Address : 0x8020 Domain0 Next Arg1 : 0x8220 Domain0 Next Mode : S-mode Domain0 SysReset : yes Boot HART ID : 2 Boot HART Domain : root Boot HART Priv Version: v1.12 Boot HART Base ISA: rv64imafdch Boot HART ISA Extensions : time,sstc Boot HART PMP Count : 16 Boot HART PMP Granularity : 4 Boot HART PMP Address Bits: 54 Boot HART MHPM Count : 16 Boot HART MIDELEG : 0x1666 Boot HART MEDELEG : 0x00f0b509 U-Boot 2023.01-rc4+dfsg-2 (Jan 06 2023 - 03:38:24 +) CPU: rv64imafdch_zicsr_zifencei_zihintpause_zba_zbb_zbc_zbs_sstc Model: riscv-virtio,qemu DRAM: 8 GiB Core: 31 devices, 15 uclasses, devicetree: board Flash: 32 MiB Loading Environment from nowhere... OK In:serial@1000 Out: serial@1000 Err: serial@1000 Net: eth0: virtio-net#2 Working FDT set to ff7344b0 Hit any key to stop autoboot: 0 Device 0: QEMU VirtIO Block Device Type: Hard Disk Capacity: 102400.0 MB = 100.0 GB (209715200 x 512) ... is now current device Scanning virtio 0:1... Found /boot/extlinux/extlinux.conf Retrieving file: /boot/extlinux/extlinux.conf U-Boot menu 1: Debian GNU/Linux bookworm/sid 6.1.0-1-riscv64 2: Debian GNU/Linux bookworm/sid 6.1.0-1-riscv64 (rescue target) 3: Debian GNU/Linux bookworm/sid 6.0.0-6-riscv64 4: Debian GNU/Linux bookworm/sid 6.0.0-6-riscv64 (rescue target) 5: Debian GNU/Linux bookworm/sid 6.0.0-5-riscv64 6: Debian GNU/Linux bookworm/sid 6.0.0-5-riscv64 (rescue target) Enter choice: 1:Debian GNU/Linux bookworm/sid 6.1.0-1-riscv64 Retrieving file: /boot/initrd.img-6.1.0-1-riscv64 Retrieving file: /boot/vmlinux-6.1.0-1-riscv64 append: root=/dev/vda1 rw noquiet Moving Image from 0x8400 to 0x8020, end=815e5000 ## Flattened Device Tree blob at ff7344b0 Booting using the fdt blob at 0xff7344b0 Working FDT set to ff7344b0 Using Device Tree in place at ff7344b0, end ff738dea Working FDT set to ff7344b0 ERROR: fdt fixup event failed: -22 - must RESET the board to recover. FDT creation failed! hanging...### ERROR ### Please RESET the board ### -8<--8<--8<--8<--8<- The source of the issue seems to be in u-boot as with u-boot-qemu 2022.10+dfsg-2 (the previous u-boot release in Debian/sid) everything works without problems (boot log below). The problem appears to be independent from the kernel image that gets booted - all three installed kernels boot fine with u-boot-qemu 2022.10+dfsg-2 but trying to boot any of them with u-boot-qemu 2023.01~rc4+dfsg-2 shows the above error. I've also tested the previous u-boot 2023.01 release candidates from experimental (2023.01-rc2+dfsg-1 and 2023.01-rc3+dfsg-1) and they show the same behaviour as 2023.01~rc4+dfsg-2. For comparison, a working boot process with u-boot-qemu 2022.10+dfsg-2:
Bug#962917: libatomic-ops: FTBFS in the testsuite on riscv64 due to missing -pthread
On Sat, Jun 20, 2020 at 12:20:14AM +0300, Ivan Maidanski wrote: > Please disregard my previous patch, and use the following one: > https://github.com/ivmai/libatomic_ops/commit/f067c258d5df3dc364857c11718be4516ca73ea0.patch > > Karsten, could you please test it? Hello, will do, but it will probably take some days. Yesterday my computer's system drive started showing massive malfunctions and I'm currently waiting for a replacement to be delivered so that I can get the system up again. Regards, Karsten -- Hiermit widerspreche ich ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#962917: libatomic-ops: FTBFS in the testsuite on riscv64 due to missing -pthread
control: severity 962917 important control: tags 962917 patch On Tue, Jun 16, 2020 at 12:40:52AM +0200, Karsten Merker wrote: > libatomic-ops FTBFS on riscv64 due to link failures in the > testsuite. The build log is available at > > > https://buildd.debian.org/status/fetch.php?pkg=libatomic-ops=riscv64=7.6.10-1=1588631725=0 > > The source of the link failures is that the tests aren't built > with the "-pthread" compiler flag. > > Some architectures such as risc64 have native atomics support, > but only for word-sized operands and not for smaller entities. > When called with "-pthread", gcc automatically links in the > necessary wrapper functions to provide the missing atomic > operations on those architectures, but this doesn't happen when > one just links in libpthread with "-lpthread". > > Building the tests with "-pthread" in CFLAGS solves the build > failures. It would be great if you could upload a new version > of the package with a corresponding change. Hello, attached is a corresponding patch. If the patch is ok for you and you would like me to perform an NMU, just let me know. Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. >From 2c4627e01637b6e2b4ad3c7d8b8ed92ff020e92b Mon Sep 17 00:00:00 2001 From: Karsten Merker Date: Wed, 17 Jun 2020 20:34:06 +0200 Subject: [PATCH] Fix FTFBS in the testsuite on riscv64 --- debian/changelog | 9 + .../libatomic-ops-enable-pthread-in-tests.diff | 12 debian/patches/series| 1 + 3 files changed, 22 insertions(+) create mode 100644 debian/patches/libatomic-ops-enable-pthread-in-tests.diff create mode 100644 debian/patches/series diff --git a/debian/changelog b/debian/changelog index 2731759..a0c6b39 100644 --- a/debian/changelog +++ b/debian/changelog @@ -1,3 +1,12 @@ +libatomic-ops (7.6.10-1.1) UNRELEASED; urgency=medium + + * Non-maintainer upload. + * Fix FTFBS in the testsuite on riscv64 by adding "-pthread" to +the testsuite's CFLAGS and thereby making sure that gcc enables +full atomics support on all platforms. (Closes: #962917) + + -- Karsten Merker Wed, 17 Jun 2020 19:37:31 +0200 + libatomic-ops (7.6.10-1) unstable; urgency=medium * New upstream release diff --git a/debian/patches/libatomic-ops-enable-pthread-in-tests.diff b/debian/patches/libatomic-ops-enable-pthread-in-tests.diff new file mode 100644 index 000..9e6891b --- /dev/null +++ b/debian/patches/libatomic-ops-enable-pthread-in-tests.diff @@ -0,0 +1,12 @@ +diff -Nur libatomic-ops-7.6.10.orig/tests/Makefile.am libatomic-ops-7.6.10/tests/Makefile.am +--- libatomic-ops-7.6.10.orig/tests/Makefile.am libatomic-ops-7.6.10/tests/Makefile.am +@@ -10,7 +10,7 @@ + -I$(top_builddir)/src -I$(top_srcdir)/src \ + -I$(top_builddir)/tests -I$(top_srcdir)/tests + +-CFLAGS += $(CFLAGS_EXTRA) ++CFLAGS += $(CFLAGS_EXTRA) -pthread + + TESTS = test_atomic$(EXEEXT) test_atomic_generalized$(EXEEXT) \ + test_stack$(EXEEXT) test_malloc$(EXEEXT) diff --git a/debian/patches/series b/debian/patches/series new file mode 100644 index 000..6bd8bef --- /dev/null +++ b/debian/patches/series @@ -0,0 +1 @@ +libatomic-ops-enable-pthread-in-tests.diff -- 2.20.1
Bug#962917: libatomic-ops: FTBFS in the testsuite on riscv64 due to missing -pthread
Source: libatomic-ops Version: 7.6.10-1 User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, libatomic-ops FTBFS on riscv64 due to link failures in the testsuite. The build log is available at https://buildd.debian.org/status/fetch.php?pkg=libatomic-ops=riscv64=7.6.10-1=1588631725=0 The source of the link failures is that the tests aren't built with the "-pthread" compiler flag. Some architectures such as risc64 have native atomics support, but only for word-sized operands and not for smaller entities. When called with "-pthread", gcc automatically links in the necessary wrapper functions to provide the missing atomic operations on those architectures, but this doesn't happen when one just links in libpthread with "-lpthread". Building the tests with "-pthread" in CFLAGS solves the build failures. It would be great if you could upload a new version of the package with a corresponding change. Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#933603: linux 5.2.1-1~exp1: riscv64 updates, including a FTBFS fix
Source: linux Severity: important Tags: patch Hello, current git master for src:linux FTBFS on riscv64. Recently CONFIG_LOAD_UEFI_KEYS has been enabled in the top-level kernel config fragment (debian/config/config), but this option depends on EFI support which is not yet available on riscv64. Therefore CONFIG_LOAD_UEFI_KEYS must be disabled in the architecture-specific config, otherwise the package fails to build on riscv64 due to undefined symbols. I have created a merge request on salsa that addresses this issue and also provides some other riscv64-related updates: [riscv64] Backport kernel image header support from kernel 5.3 [riscv64] Enable clock drivers for the SiFive FU540 [riscv64] Enable SiFive UART and UART console support [riscv64] Disable CONFIG_LOAD_UEFI_KEYS for riscv64 (fixes FTBFS) https://salsa.debian.org/kernel-team/linux/merge_requests/161/commits Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#928735: u-boot-menu: Avoid hard-coding "vmlinuz"
On Thu, 09 May 2019, Vagrant Cascadian wrote: > Package: u-boot-menu > Severity: normal > Version: 3 > Tags: patch > > Some architectures which might make use of u-boot-menu do not have > kernel files matching "vmlinuz" (e.g. riscv64 has "vmlinux"). > > The attached patch uses "linux-version list --paths" to output the path > of each versioned kernel, which outputs the matching kernel filenames. Hello, may I kindly ping you on this bugreport? I'm working on d-i support for riscv64 and would like to use the u-boot-menu package there, but without Vagrant's patch it isn't usable on riscv64 (and on quite a number of other architectures as well, please see below). It would be great if you could upload a new version of u-boot-menu with the patch applied to unstable. If you are ok with the patch but don't have time for preparing an upload, I would offer doing an NMU if that would be ok for you. An overview of vmlinuz- vs. vmlinux-using Debian architectures: vmlinuz: alpha, amd64, arm64, armel, armhf, hppa, i386, ia64, s390x, sh4 vmlinux: m68k, mips{,64}{,r6}{,el}, powerpc, powerpcspe, ppc64, ppc64el, riscv64, sparc64 Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#928832: elfutils: diff for NMU version 0.176-1.1
Control: tags 928832 + pending Dear maintainer, as previously discussed I've prepared an NMU for elfutils (versioned as 0.176-1.1) and will upload it shortly. Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nru elfutils-0.176/debian/changelog elfutils-0.176/debian/changelog --- elfutils-0.176/debian/changelog 2019-02-16 14:54:50.0 +0100 +++ elfutils-0.176/debian/changelog 2019-05-28 20:53:12.0 +0200 @@ -1,3 +1,10 @@ +elfutils (0.176-1.1) unstable; urgency=medium + + * Non-maintainer upload with maintainer permission + * Fixes FTBFS on riscv64 (Closes: #928832) + + -- Karsten Merker Tue, 28 May 2019 20:53:12 +0200 + elfutils (0.176-1) unstable; urgency=medium * New upstream release diff -Nru elfutils-0.176/debian/control elfutils-0.176/debian/control --- elfutils-0.176/debian/control 2018-06-24 20:54:48.0 +0200 +++ elfutils-0.176/debian/control 2019-05-28 20:52:01.0 +0200 @@ -1,7 +1,7 @@ Source: elfutils Priority: optional Maintainer: Kurt Roeckx -Build-Depends: debhelper (>= 9), autotools-dev, autoconf, automake, bzip2, zlib1g-dev, zlib1g-dev:native, libbz2-dev, liblzma-dev, m4, gettext, autoconf, automake, gawk, dpkg-dev (>= 1.16.1~), gcc-multilib [any-amd64 sparc64] , libc6-dbg [powerpc powerpcspe ppc64 ppc64el armel armhf arm64 sparc64], flex, bison +Build-Depends: debhelper (>= 9), autotools-dev, autoconf, automake, bzip2, zlib1g-dev, zlib1g-dev:native, libbz2-dev, liblzma-dev, m4, gettext, autoconf, automake, gawk, dpkg-dev (>= 1.16.1~), gcc-multilib [any-amd64 sparc64] , libc6-dbg [powerpc powerpcspe ppc64 ppc64el armel armhf arm64 sparc64 riscv64], flex, bison Build-Conflicts: autoconf2.13, automake1.4 Standards-Version: 4.1.0 Section: libs diff -Nru elfutils-0.176/debian/patches/riscv-retval-workaround.patch elfutils-0.176/debian/patches/riscv-retval-workaround.patch --- elfutils-0.176/debian/patches/riscv-retval-workaround.patch 1970-01-01 01:00:00.0 +0100 +++ elfutils-0.176/debian/patches/riscv-retval-workaround.patch 2019-05-28 20:52:01.0 +0200 @@ -0,0 +1,11 @@ +--- a/backends/riscv_retval.c b/backends/riscv_retval.c +@@ -111,7 +111,7 @@ + Dwarf_Die *arg1 __attribute__ ((unused))) + { + /* ??? */ +- return 1; ++ return 0; + } + + static int diff -Nru elfutils-0.176/debian/patches/series elfutils-0.176/debian/patches/series --- elfutils-0.176/debian/patches/series 2019-02-16 14:54:50.0 +0100 +++ elfutils-0.176/debian/patches/series 2019-05-28 20:52:01.0 +0200 @@ -11,3 +11,4 @@ ignore_strmerge.diff disable_werror.patch mips_cfi.patch +riscv-retval-workaround.patch
Bug#928832: [PATCH] elfutils: FTBFS on riscv64
On Sat, May 11, 2019 at 10:26:37PM +0200, Karsten Merker wrote: > elfutils currently FTBFS for riscv64 and this is the last remaining > blocker to get a working debian-installer for riscv64. As already > discussed together with elfutils upstream, there is a workaround that > requires only a one-line change in a riscv-specific file plus > extending the existing architecture-specific build-dependency on > libc6-dbg for [powerpc powerpcspe ppc64 ppc64el armel armhf arm64 > sparc64] to also include riscv64. > > I would therefore like to ask you to upload a version of elfutils > with the attached patch to unstable. The patch should pose no > problem with the freeze as it only touches an arch-specific source > file for a non-release architecture and doesn't have any effect at > all on any release architecture. At the same time it is very > important for the riscv64 port as without the patch we cannot make > progress with d-i on riscv64, and having a working d-i is a > significant milestone for the riscv64 port. Hello Kurt, may I kindly ping you on this bugreport? Is there anything that speaks against application of the patch? If you think that the patch is ok but you don't have time for preparing a new upload, I would offer doing an NMU if that would be ok for you. Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#928832: [PATCH] elfutils: FTBFS on riscv64
Source: elfutils Version: 0.176-1 Severity: important Tags: patch User: debian-ri...@lists.debian.org Usertags: riscv64 Hello Kurt, elfutils currently FTBFS for riscv64 and this is the last remaining blocker to get a working debian-installer for riscv64. As already discussed together with elfutils upstream, there is a workaround that requires only a one-line change in a riscv-specific file plus extending the existing architecture-specific build-dependency on libc6-dbg for [powerpc powerpcspe ppc64 ppc64el armel armhf arm64 sparc64] to also include riscv64. I would therefore like to ask you to upload a version of elfutils with the attached patch to unstable. The patch should pose no problem with the freeze as it only touches an arch-specific source file for a non-release architecture and doesn't have any effect at all on any release architecture. At the same time it is very important for the riscv64 port as without the patch we cannot make progress with d-i on riscv64, and having a working d-i is a significant milestone for the riscv64 port. Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur elfutils-0.176/debian/control elfutils-0.176.patched/debian/control --- elfutils-0.176/debian/control +++ elfutils-0.176.patched/debian/control @@ -1,7 +1,7 @@ Source: elfutils Priority: optional Maintainer: Kurt Roeckx -Build-Depends: debhelper (>= 9), autotools-dev, autoconf, automake, bzip2, zlib1g-dev, zlib1g-dev:native, libbz2-dev, liblzma-dev, m4, gettext, autoconf, automake, gawk, dpkg-dev (>= 1.16.1~), gcc-multilib [any-amd64 sparc64] , libc6-dbg [powerpc powerpcspe ppc64 ppc64el armel armhf arm64 sparc64], flex, bison +Build-Depends: debhelper (>= 9), autotools-dev, autoconf, automake, bzip2, zlib1g-dev, zlib1g-dev:native, libbz2-dev, liblzma-dev, m4, gettext, autoconf, automake, gawk, dpkg-dev (>= 1.16.1~), gcc-multilib [any-amd64 sparc64] , libc6-dbg [powerpc powerpcspe ppc64 ppc64el armel armhf arm64 sparc64 riscv64], flex, bison Build-Conflicts: autoconf2.13, automake1.4 Standards-Version: 4.1.0 Section: libs diff -Nur elfutils-0.176/debian/patches/riscv-retval-workaround.patch elfutils-0.176.patched/debian/patches/riscv-retval-workaround.patch --- elfutils-0.176/debian/patches/riscv-retval-workaround.patch 1970-01-01 01:00:00.0 +0100 +++ elfutils-0.176.patched/debian/patches/riscv-retval-workaround.patch 2019-05-11 20:26:06.736089789 +0200 @@ -0,0 +1,11 @@ +--- a/backends/riscv_retval.c b/backends/riscv_retval.c +@@ -111,7 +111,7 @@ + Dwarf_Die *arg1 __attribute__ ((unused))) + { + /* ??? */ +- return 1; ++ return 0; + } + + static int diff -Nur elfutils-0.176/debian/patches/series elfutils-0.176.patched/debian/patches/series --- elfutils-0.176/debian/patches/series +++ elfutils-0.176.patched/debian/patches/series @@ -11,3 +11,4 @@ ignore_strmerge.diff disable_werror.patch mips_cfi.patch +riscv-retval-workaround.patch
Bug#928451: linux: riscv64 updates (change kernel image type to flat image and enable vdso)
ting OpenNTPd Network Time Protocol... [ OK ] Started OpenNTPd Network Time Protocol. [ OK ] Started Raise network interfaces. Starting Online ext4 Metad…a Check for All Filesystems... Starting Rotate log files... Starting Daily man-db regeneration... [ OK ] Reached target Network. Starting Permit User Sessions... Starting OpenBSD Secure Shell server... [ OK ] Reached target Network is Online. Starting LSB: exim Mail Transport Agent... Starting Daily apt download activities... [ OK ] Started Permit User Sessions. [ OK ] Started Serial Getty on ttyS0. [ OK ] Started Getty on tty1. [ OK ] Reached target Login Prompts. [ OK ] Started OpenBSD Secure Shell server. [ OK ] Started Online ext4 Metadata Check for All Filesystems. [ OK ] Started Rotate log files. [ OK ] Started Recover schroot sessions. Debian GNU/Linux 10 riscv64 ttyS0 riscv64 login: -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. >From de70ae43563da85568562e49c6f3bc1f27741dfc Mon Sep 17 00:00:00 2001 From: Karsten Merker Date: Sat, 4 May 2019 19:41:45 +0200 Subject: [PATCH 1/2] [riscv64] Change the kernel image format from ELF to flat Image. The riscv64 architecture is changing its standard kernel image format from ELF to a flat kernel image with a PE/COFF-compatible header (similar to arm64) to make EFI stub support possible. Ship arch/riscv/boot/Image instead of an ELF vmlinux in accordance with this change. --- debian/config/riscv64/defines | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/debian/config/riscv64/defines b/debian/config/riscv64/defines index 2ea6d75977f1..e83b750ea74a 100644 --- a/debian/config/riscv64/defines +++ b/debian/config/riscv64/defines @@ -4,7 +4,7 @@ featuresets: none [build] -image-file: vmlinux +image-file: arch/riscv/boot/Image [image] install-stem: vmlinux -- 2.20.1 >From 5fda00263bfe81d8a62dd4162c66527b51ea1755 Mon Sep 17 00:00:00 2001 From: Karsten Merker Date: Sat, 4 May 2019 19:51:14 +0200 Subject: [PATCH 2/2] [riscv64] Enable vdso When riscv64 support was originally added to the Debian Linux kernel package, the mainline kernel lacked a vdso_install target for riscv64. This has in the meantime been added with upstream commit f157d411a9eb170d2ee6b766da7a381962017cc9 ("riscv: add missing vdso_install target"), so we can now enable the corresponding option in the kernel package. --- debian/config/riscv64/defines | 1 + 1 file changed, 1 insertion(+) diff --git a/debian/config/riscv64/defines b/debian/config/riscv64/defines index e83b750ea74a..1d5a58b7313a 100644 --- a/debian/config/riscv64/defines +++ b/debian/config/riscv64/defines @@ -5,6 +5,7 @@ featuresets: [build] image-file: arch/riscv/boot/Image +vdso: true [image] install-stem: vmlinux -- 2.20.1
Bug#928079: mmdebstrap: fails when passing --include="linux-image-amd64"
Package: mmdebstrap Version: 0.4.1-2 Hello, when trying to include a kernel image in the package list for mmdebstrap, mmdebstrap fails when running the deferred trigger for update-initramfs. The effect is independent of the architecture (tested on amd64, armhf and riscv64). When doing a standard install without the "--include" parameter, switching to the chroot and manually installing the same kernel package everything works without problems. Following are three logs: - mmdebstrap called with --include="linux-image-amd64" - mmdebstrap called without --include and manual kernel installation afterwards - same as before but trying to (at least partially) replicate the apt environment in mmdebstrap Failure log === $ sudo mmdebstrap --aptopt='Acquire::http { Proxy "http://127.0.0.1:3142;; }' --architectures=amd64 --include="linux-image-amd64" sid /tmp/amd64-chroot "deb http://deb.debian.org/debian/ sid main" I: automatically chosen mode: root I: chroot architecture amd64 is equal to the host's architecture I: running apt-get update... done I: downloading packages with apt... done I: extracting archives... done I: installing packages... done I: downloading apt... done I: installing apt... done I: installing remaining packages inside the chroot... done Reading package lists... Building dependency tree... adduser is already the newest version (3.118). apt is already the newest version (1.8.0). debconf is already the newest version (1.5.71). debian-archive-keyring is already the newest version (2019.1). gpgv is already the newest version (2.2.13-1). mawk is already the newest version (1.3.3-17+b3). libpam-modules is already the newest version (1.3.1-5). libpam-modules-bin is already the newest version (1.3.1-5). libpam-runtime is already the newest version (1.3.1-5). passwd is already the newest version (1:4.5-1.1). fdisk is already the newest version (2.33.1-0.1). The following additional packages will be installed: dmsetup initramfs-tools initramfs-tools-core klibc-utils libapparmor1 libapt-inst2.0 libargon2-1 libbsd0 libcap2 libcap2-bin libcom-err2 libcryptsetup12 libdevmapper1.02.1 libdns-export1104 libelf1 libestr0 libext2fs2 libfastjson4 libidn11 libip4tc0 libip6tc0 libiptc0 libisc-export1100 libjson-c3 libklibc libkmod2 liblocale-gettext-perl liblognorm5 libmnl0 libncurses6 libnetfilter-conntrack3 libnewt0.52 libnfnetlink0 libnftnl11 libpopt0 libprocps7 libslang2 libss2 libssl1.1 libtext-charwidth-perl libtext-iconv-perl libtext-wrapi18n-perl libxtables12 linux-base linux-image-4.19.0-4-amd64 lsb-base xxd Suggested packages: cpp wamerican | wordlist whois vacation libarchive1 anacron checksecurity gpart parted fuse2fs e2fsck-static ppp rdnssd bash-completion iproute2-doc resolvconf avahi-autoipd isc-dhcp-client-ddns linux-doc-4.19 debian-kernel-handbook grub-pc | grub-efi-amd64 | extlinux nfs-common spell readline-doc rsyslog-mysql | rsyslog-pgsql rsyslog-mongodb rsyslog-doc rsyslog-gnutls rsyslog-gssapi rsyslog-relp systemd-container policykit-1 indent Recommended packages: default-mta | mail-transport-agent e2fsprogs-l10n lvm2 busybox | busybox-static pigz libatm1 nftables libpam-cap libgpm2 libfribidi0 firmware-linux-free irqbalance apparmor bsd-mailx | mailx psmisc libpam-systemd dbus libnss-systemd laptop-detect The following NEW packages will be installed: apt-utils bsdmainutils cpio cron debconf-i18n dmidecode dmsetup e2fsprogs gdbm-l10n ifupdown init initramfs-tools initramfs-tools-core iproute2 iptables iputils-ping isc-dhcp-client isc-dhcp-common klibc-utils kmod less libapparmor1 libapt-inst2.0 libargon2-1 libbsd0 libcap2 libcap2-bin libcom-err2 libcryptsetup12 libdevmapper1.02.1 libdns-export1104 libelf1 libestr0 libext2fs2 libfastjson4 libidn11 libip4tc0 libip6tc0 libiptc0 libisc-export1100 libjson-c3 libklibc libkmod2 liblocale-gettext-perl liblognorm5 libmnl0 libncurses6 libnetfilter-conntrack3 libnewt0.52 libnfnetlink0 libnftnl11 libpopt0 libprocps7 libslang2 libss2 libssl1.1 libtext-charwidth-perl libtext-iconv-perl libtext-wrapi18n-perl libxtables12 linux-base linux-image-4.19.0-4-amd64 linux-image-amd64 logrotate lsb-base mount nano netbase procps readline-common rsyslog sensible-utils systemd systemd-sysv tasksel tasksel-data tzdata udev vim-common vim-tiny whiptail xxd 0 upgraded, 82 newly installed, 0 to remove and 0 not upgraded. Need to get 66.0 MB of archives. After this operation, 334 MB of additional disk space will be used. Get:1 http://deb.debian.org/debian sid/main amd64 libcom-err2 amd64 1.45.0-1 [69.1 kB] Get:2 http://deb.debian.org/debian sid/main amd64 libext2fs2 amd64 1.45.0-1 [256 kB] Get:3 http://deb.debian.org/debian sid/main amd64 libss2 amd64 1.45.0-1 [73.6 kB] Get:4 http://deb.debian.org/debian sid/main amd64 e2fsprogs amd64 1.45.0-1 [592 kB] Get:5 http://deb.debian.org/debian sid/main amd64 mount amd64 2.33.1-0.1 [174 kB] Get:6 http://deb.debian.org/debian sid/main amd64
Bug#927744: debmake: Please add support for SPDX-License-Identifiers
Package: debmake Version: 4.3.1-1 Severity: wishlist Hello, more and more upstream projects (including a number of high-profile ones such as the Linux kernel) are moving to including SPDX license identifiers (https://spdx.org/ids) instead of the full license text (in case of BSD- and MIT-style licenses) respectively the short-form (L)GPL notice in their source files. These identifiers are intended to make the license a piece of code is distributed under easily machine-parsable while keeping them human-readable. Examples: - SPDX-License-Identifier: BSD-2-Clause SPDX-License-Identifier: GPL-2.0 SPDX-License-Identifier: GPL-2.0-or-later Currently "debmake -cc" doesn't recognize the SPDX IDs and flags all files using them as ""License: __UNKNOWN__". It would be great if you could extend the patterns that "debmake -cc" matches on for recognizing the various licenses to include the corresponding SPDX license identifiers. A list of the identifiers and the corresponding full license text is available at https://spdx.org/licenses/. Regards, Karsten -- Ich widerspreche hiermit ausdrücklich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#918428: choose-mirror: mirror list generation broken for all ports architectures
Source: choose-mirror Version: 2.96 Hello, while working on riscv64 support in d-i I have stumbled over a problem with choose-mirror: instead of providing a mirror list, choose-mirror only offers the option "enter information manually", even if Mirrors.masterlist is extended to contain a "riscv64" entry in the "Ports-architecture" field for deb.debian.org and ftp.ports.debian.org. I have tracked the source of this issue down to the part of the "mirrorlist" script that tries to filter out mirrors that don't carry the architecture for which choose-mirror is currently built: https://salsa.debian.org/installer-team/choose-mirror/blob/8bc40f7e97afa5fb075535f501abc31614a574dd/mirrorlist#L107 This filter function only takes the "Archive-architecture" fields from Mirrors.masterlist into account, but not the "Ports-architecture" fields, so it always results in an empty list for Debian-Ports architectures. Unfortunately I don't really speak perl (just enough to have a rough idea of what the mirrorlist script does, but not enough to make modifications to it with good conscience), so it would be great if somebody from debian-boot with better perl knowledge than me could take a look. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#908161: Please enable building a riscv64 kernel image
Control: tags 908161 + patch On Tue, Sep 18, 2018 at 08:57:01PM +0200, Karsten Merker wrote: > On Sat, Sep 08, 2018 at 11:15:36PM +0100, Ben Hutchings wrote: > > [Building a linux-image-*-riscv64 binary package] > > > The addition of riscv will have to wait until it has support > > for an initramfs. > > > > Is this commit sufficient to make booting with an initramfs work: > > > > commit cdc7274029ca5984350a057a2399aaa340d3be2d > > Author: Guenter Roeck > > Date: Tue Aug 28 17:33:46 2018 -0700 > > > > riscv: Do not overwrite initrd_start and initrd_end > > > > or are more changes needed? > > Hello, > > just a short status update: the aforementioned patch has been > included in the upstream 4.19-rc4 release and I can confirm > that the initramfs support for riscv64 works with 4.19-rc4. > > The other major issue in this bug (unversioned symbols breaking > the package build) is still unresolved; I'll report back as soon > as I have received feedback from the upstream RISC-V architecture > maintainer. Hello, all previously mentioned issues have been addressed in the meantime: - The broken initrd support has been fixed upstream in kernel 4.19-rc4. - The symbol version issue has been fixed upstream in kernel 4.19-rc6. - The riscv64 kernel config has been modularized as far as possible and all redundant entries have been removed. - Headings have been added to the kernel config. Attached is a new patch, alternatively it is available as a merge request on salsa as suggested earlier in the discussion: https://salsa.debian.org/kernel-team/linux/merge_requests/66 The resulting kernel has been successfully tested on a qemu "virt" board: [0.00] OF: fdt: Ignoring memory range 0x8000 - 0x8020 [0.00] Linux version 4.19.0-rc7-riscv64 (debian-ker...@lists.debian.org) (gcc version 8.2.0 (Debian 8.2.0-7)) #1 SMP Debian 4.19~rc7-1~exp2 (2018-10-08) [0.00] bootconsole [early0] enabled [0.00] Initial ramdisk at: 0x(ptrval) (43521258 bytes) [0.00] Zone ranges: [0.00] DMA32[mem 0x8020-0x] [0.00] Normal [mem 0x0001-0x2fff] [0.00] Movable zone start for each node [0.00] Early memory node ranges [0.00] node 0: [mem 0x8020-0x0002] [0.00] Initmem setup node 0 [mem 0x8020-0x0002] [0.00] On node 0 totalpages: 2620928 [0.00] DMA32 zone: 8184 pages used for memmap [0.00] DMA32 zone: 0 pages reserved [0.00] DMA32 zone: 523776 pages, LIFO batch:63 [0.00] Normal zone: 32768 pages used for memmap [0.00] Normal zone: 2097152 pages, LIFO batch:63 [0.00] software IO TLB: mapped [mem 0xfbfff000-0xf000] (64MB) [0.00] elf_hwcap is 0x112d [0.00] percpu: Embedded 19 pages/cpu @(ptrval) s39384 r8192 d30248 u77824 [0.00] pcpu-alloc: s39384 r8192 d30248 u77824 alloc=19*4096 [0.00] pcpu-alloc: [0] 0 [0] 1 [0] 2 [0] 3 [0.00] Built 1 zonelists, mobility grouping on. Total pages: 2579976 [0.00] Kernel command line: console=ttyS0 ro root=/dev/vda [0.00] Dentry cache hash table entries: 2097152 (order: 12, 16777216 bytes) [0.00] Inode-cache hash table entries: 1048576 (order: 11, 8388608 bytes) [0.00] Sorting __ex_table... [0.00] Memory: 10178016K/10483712K available (4955K kernel code, 504K rwdata, 1633K rodata, 446K init, 934K bss, 305696K reserved, 0K cma-reserved) [0.00] random: get_random_u64 called from __kmem_cache_create+0x46/0x55c with crng_init=0 [0.00] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=4, Nodes=1 [0.00] ftrace: allocating 21055 entries in 83 pages [0.00] rcu: Hierarchical RCU implementation. [0.00] rcu: RCU restricting CPUs from NR_CPUS=8 to nr_cpu_ids=4. [0.00] rcu: Adjusting geometry for rcu_fanout_leaf=16, nr_cpu_ids=4 [0.00] NR_IRQS: 0, nr_irqs: 0, preallocated irqs: 0 [0.00] plic: mapped 10 interrupts to 4 (out of 8) handlers. [0.00] clocksource: riscv_clocksource: mask: 0x max_cycles: 0x24e6a1710, max_idle_ns: 440795202120 ns [0.004000] Console: colour dummy device 80x25 [0.008000] Calibrating delay loop (skipped), value calculated using timer frequency.. 20.00 BogoMIPS (lpj=4) [0.012000] pid_max: default: 32768 minimum: 301 [0.016000] Security Framework initialized [0.016000] Yama: disabled by default; enable with sysctl kernel.yama.* [0.02] AppArmor: AppArmor initialized [0.024000] Mount-cache hash table entries: 32768 (order: 6, 262144 bytes) [0.028000] Mountpoint-cache hash table entries: 32768 (order: 6, 262144 bytes) [0.072000] rcu: Hierarchical SRCU
Bug#910044: mmdebstrap: problem with sources.list-style mirror parameter handling
On Tue, Oct 02, 2018 at 10:13:42AM +0200, Johannes Schauer wrote: > Quoting Johannes Schauer (2018-10-02 04:23:54) > > > Another thing that I have stumbled upon is that providing a sources.list > > > on > > > stdin only works when no target directory is provided on the commandline, > > > i.e. when automatically creating a tarball. While > > > > > > echo -e "deb http://deb.debian.org/debian-ports/ sid main\ndeb > > > http://deb.debian.org/debian-ports/ unreleased main" | mmdebstrap > > > --architectures=riscv64 > /tmp/rv64-chroot.tar > > > > > > works, the following doesn't: > > > > > > echo -e "deb http://deb.debian.org/debian-ports/ sid main\ndeb > > > http://deb.debian.org/debian-ports/ unreleased main" | mmdebstrap > > > --architectures=riscv64 sid /tmp/rv64-chroot > > > > > > It results in > > > > > > I: riscv64 cannot be executed, falling back to qemu-user > > > I: automatically chosen mode: root > > > I: running apt-get update... > > > done > > > apt-get update didn't download anything at /usr/bin/mmdebstrap line 721. > > > > This is actually a feature. In your second invocation, you didn't specify > > the > > "MIRROR" argument. The docs say: > > > > > If no MIRROR option is provided, http://deb.debian.org/debian is used. > > > > In your first invocation, you also didn't specify the "SUITE" argument. The > > docs say: > > > > > If no SUITE was specified, then a single MIRROR "-" is added and thus the > > > information of the desired suite has to come from standard input as part > > > of a > > > valid apt sources.list file. > > > > So the implicit '-' as the mirror argument only works if there was no > > SUITE. If > > you pass a SUITE, you need to explicitly tell it that you want to read the > > mirror from stdin. This is to have compatibility with how debootstrap also > > just > > uses deb.debian.org/debian as the mirror if the mirror argument is missing. > > > > To make the cause of the problem more obvious, I'm now printing the content > > of > > /etc/apt/sources.list in the chroot in case apt-get didn't download > > anything: > > > > https://gitlab.mister-muffin.de/josch/mmdebstrap/commit/1f13d0157bc8364ac491203a6a9156a78f6228a9 > > > > But we could turn this bug into a feature request of the following form: > > > > If no MIRROR was provided *and* something was specified on standard input, > > then > > don't use deb.debian.org/debian but use whatever was given on standard input > > instead. > > > > Or even more liberal: > > > > If anything is provided on standard input and there is no '-' MIRROR > > argument, > > then just append whatever is given on standard input to the apt sources.list > > inside the chroot. > > > > What do you think? > > now that I've slept about it, I think your idea makes sense. If you don't > specify a mirror but you pass something on standard input, then the default > mirror should *not* be used. The default mirror should only be used if neither > a mirror nor anything from standard input was given. > > I pushed a fix to the upstream repository. I would be happy if you could > download the latest version of the shell script from there and test if it > works > as you had expected it to work. Many thanks! I've run a number of tests covering different options (native/foreign chroot, implicit/explicit mirror, plain-url/sources.list-entry, mirror information via stdin/on the commandline, target directory/tarball) with the current upstream git head and everything looks good: * native chroot with default mirror and with a directory as the target: # mmdebstrap --include="debian-ports-archive-keyring" sid /tmp/amd64-chroot I: chroot architecture amd64 is equal to the host's architecture I: automatically chosen mode: root I: running apt-get update... done I: downloading packages with apt... done I: extracting archives... done I: installing packages... done I: downloading apt... done I: installing apt... done I: installing remaining packages inside the chroot... done I: cleaning package lists and apt cache... done done * native chroot with a mirror URL on the commandline and a directory as the target: # mmdebstrap --include="debian-ports-archive-keyring" sid /tmp/amd64-chroot http://deb.debian.org/debian I: chroot architecture amd64 is equal to the host's architecture I: automatically chosen mode: root I: running apt-get update... done I: downloading packages with apt... done I: extracting archives... done I: installing packages... done I: downloading apt... done I: installing apt... done I: installing remaining packages inside the chroot... done I: cleaning package lists and apt cache... done done * foreign (riscv64) chroot with sources.list passed on stdin and with a directory as the target: # echo -e "deb http://deb.debian.org/debian-ports/ sid main\ndeb http://deb.debian.org/debian-ports/ unreleased main" | mmdebstrap
Bug#910044: mmdebstrap: problem with sources.list-style mirror parameter handling
Package: mmdebstrap Version: 0.1.0-2 Hello, many thanks for creating mmdebstrap! I'm using it for creating chroots for Debian-Ports architectures, where we need to pull the packages from two suites: unstable and unreleased. While using mmdebstrap, I've stumbled over two issues. Everything works fine when creating a sources.list, feeding it to mmdebstrap on stdin and having mmdebstrap automatically generate a tarball, i.e. with something like echo -e "deb http://deb.debian.org/debian-ports/ sid main\ndeb http://deb.debian.org/debian-ports/ unreleased main" | mmdebstrap --architectures=riscv64 > /tmp/rv64-chroot.tar The process fails when one tries to pass the mirrors on the commandline, though. Due to the fact that there is more than one suite to use, the mirrors need to be passed in sources.list style, i.e. as "deb http:///debian-ports main". Running mmdebstrap with mmdebstrap --architectures=riscv64 sid /tmp/rv64-chroot "deb http://deb.debian.org/debian-ports/ sid main" "deb http://deb.debian.org/debian-ports/ unreleased main" results in E: Malformed entry 1 in list file /tmp/rv64-chroot/etc/apt/sources.list (URI parse) E: The list of sources could not be read. apt-get update failed at /usr/bin/mmdebstrap line 548. # cat /tmp/rv64-chroot/etc/apt/sources.list deb deb http://deb.debian.org/debian-ports/ sid main sid main deb deb http://deb.debian.org/debian-ports/ unreleased main sid main The manpage states: If a MIRROR option starts with "deb " or "deb-src " then it is used as a one-line-style format entry for apt's sources.list inside the chroot. If a MIRROR option contains a "://" then it is interpreted as a mirror URI and the apt line inside the chroot is assembled as "deb [arch=A] B C D" where A is the host's native architecture, B is the MIRROR, C is the given SUITE and D is the components given via --components (defaults to "main"). It looks like the second part is applied unconditionally on all mirror parameters that contain a "://", although that should not happen if the mirror parameter starts with a "deb " or "deb-src " and thereby already constitutes a complete sources.list entry. Another thing that I have stumbled upon is that providing a sources.list on stdin only works when no target directory is provided on the commandline, i.e. when automatically creating a tarball. While echo -e "deb http://deb.debian.org/debian-ports/ sid main\ndeb http://deb.debian.org/debian-ports/ unreleased main" | mmdebstrap --architectures=riscv64 > /tmp/rv64-chroot.tar works, the following doesn't: echo -e "deb http://deb.debian.org/debian-ports/ sid main\ndeb http://deb.debian.org/debian-ports/ unreleased main" | mmdebstrap --architectures=riscv64 sid /tmp/rv64-chroot It results in I: riscv64 cannot be executed, falling back to qemu-user I: automatically chosen mode: root I: running apt-get update... done apt-get update didn't download anything at /usr/bin/mmdebstrap line 721. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#908161: Please enable building a riscv64 kernel image
On Thu, Sep 06, 2018 at 10:28:19PM +0100, Ben Hutchings wrote: > On Thu, 2018-09-06 at 22:06 +0200, Karsten Merker wrote: > > Source: linux > > Version: 4.19~rc2-1~exp1 > > Severity: wishlist [...] > > starting with version 4.19rc2, the mainline Linux kernel includes > > all drivers necessary for running a riscv64 system in qemu, so it > > would be great if the "linux" source package could be extended to > > build a linux-image-*-riscv64 binary package. > > > > Attached is a patch that tries to add the necessary bits. > > This config sets a whole lot of things to be built-in, but our policy > is to build everything as modules if it works properly work as a > module. This will also cause the building of installer udebs to fail > (empty packages are treated as a fatal error). Hello, the reason for using a static config was that using an initrd isn't possible on riscv64 with kernel 4.19rc2. This will hopefully change sometime before the final 4.19 release so that we can move to a fully modularized config, but for now everyting required to mount the rootfs and bring up init has to be built-in. I can probably trim down the current static config a bit more, but e.g. filesystem drivers need to be built-in for now, otherwise mounting the rootfs isn't possible. > It also seems to have some redundant settings. debian/config/config is > always used first (see README.source), so don't repeat anything that's > in there. Many thanks for the pointer, I'll take that into account for the next version of the patch. > Finally you should use kconfigeditor2 to add headings to the config > file. You need to check out the kernel-team.git repository, and then > in the linux repository run something like: > > ../kernel-team/utils/kconfigeditor2/process.py . Ok, will do. > > Unfortunately, with the patch applied the kernel itself builds > > successfully, but the package build process then fails with > > > > -8<--8<--8<--8<--8<- > > > > make[3]: Leaving directory > > '<>/linux-4.19~rc2/debian/build/build_riscv64_none_riscv64' > > debian/bin/buildcheck.py debian/build/build_riscv64_none_riscv64 riscv64 > > none riscv64 > > ABI is not completely versioned! Refusing to continue. > > > > Unversioned symbols: > > _mcount module: vmlinux, version: > > 0x, export: EXPORT_SYMBOL > > return_to_handlermodule: vmlinux, version: > > 0x, export: EXPORT_SYMBOL > > Can't read ABI reference. ABI not checked! > > make[2]: *** [debian/rules.real:217: > > debian/stamps/build_riscv64_none_riscv64] Fehler 1 > > > > -8<--8<--8<--8<--8<- > > > > I'm somewhat stuck here - is this an upstream issue or > > have I overlooked something on the packaging side? Pointers > > welcome :). > > It's an upstream issue, but not a fatal error there. For Debian it is > a fatal error becasue unversioned symbols potentially undermine code > signing. > > Any symbol exported from an assembly-language file won't automatically > get a symbol version, since there's no type information there. The way > to fix this is to include (or directly) add the function prototypes in > arch/riscv/include/asm/asm-prototypes.h. > > I don't think that return_to_handler should be exported at all. No > other architecture does. As for _mcount, that is declared in > , so should just be: > > /* SPDX-License-Identifier: GPL-2.0 */ > #include Thanks for the explanation, I'll contact the upstream RISC-V kernel maintainer regarding this. > Finally, you have added module lists for installer udebs, but this > won't have any effect unless you also add the new architecture and > flavour to debian/installer/kernel-versions. Again thanks for the pointer, I'll look into it. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#908161: Please enable building a riscv64 kernel image
Source: linux Version: 4.19~rc2-1~exp1 Severity: wishlist Tags: patch X-Debbugs-CC: debian-ri...@lists.debian.org User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, starting with version 4.19rc2, the mainline Linux kernel includes all drivers necessary for running a riscv64 system in qemu, so it would be great if the "linux" source package could be extended to build a linux-image-*-riscv64 binary package. Attached is a patch that tries to add the necessary bits. Unfortunately, with the patch applied the kernel itself builds successfully, but the package build process then fails with -8<--8<--8<--8<--8<- make[3]: Leaving directory '<>/linux-4.19~rc2/debian/build/build_riscv64_none_riscv64' debian/bin/buildcheck.py debian/build/build_riscv64_none_riscv64 riscv64 none riscv64 ABI is not completely versioned! Refusing to continue. Unversioned symbols: _mcount module: vmlinux, version: 0x, export: EXPORT_SYMBOL return_to_handlermodule: vmlinux, version: 0x, export: EXPORT_SYMBOL Can't read ABI reference. ABI not checked! make[2]: *** [debian/rules.real:217: debian/stamps/build_riscv64_none_riscv64] Fehler 1 -8<--8<--8<--8<--8<- I'm somewhat stuck here - is this an upstream issue or have I overlooked something on the packaging side? Pointers welcome :). Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. >From d1d47c22d5c41bb3b1924d2534aa06e86a16c10d Mon Sep 17 00:00:00 2001 From: Karsten Merker Date: Sat, 1 Sep 2018 23:02:11 +0200 Subject: [PATCH] Build a kernel image for riscv64 --- debian/config/riscv64/config | 79 +++ debian/config/riscv64/defines | 6 +- debian/config/riscv64/none/defines| 3 + debian/installer/modules/riscv64/ata-modules | 1 + .../installer/modules/riscv64/btrfs-modules | 1 + .../modules/riscv64/compress-modules | 1 + debian/installer/modules/riscv64/crc-modules | 1 + .../modules/riscv64/crypto-dm-modules | 1 + .../installer/modules/riscv64/crypto-modules | 1 + .../installer/modules/riscv64/event-modules | 1 + debian/installer/modules/riscv64/ext4-modules | 1 + debian/installer/modules/riscv64/fat-modules | 1 + debian/installer/modules/riscv64/fuse-modules | 1 + debian/installer/modules/riscv64/i2c-modules | 1 + .../installer/modules/riscv64/input-modules | 1 + .../installer/modules/riscv64/isofs-modules | 1 + debian/installer/modules/riscv64/jfs-modules | 1 + debian/installer/modules/riscv64/kernel-image | 1 + debian/installer/modules/riscv64/leds-modules | 1 + debian/installer/modules/riscv64/loop-modules | 1 + debian/installer/modules/riscv64/md-modules | 1 + debian/installer/modules/riscv64/mmc-modules | 1 + debian/installer/modules/riscv64/mtd-modules | 1 + .../modules/riscv64/multipath-modules | 1 + debian/installer/modules/riscv64/nbd-modules | 1 + debian/installer/modules/riscv64/nic-modules | 1 + .../modules/riscv64/nic-shared-modules| 1 + .../installer/modules/riscv64/nic-usb-modules | 1 + .../modules/riscv64/nic-wireless-modules | 1 + debian/installer/modules/riscv64/pata-modules | 1 + debian/installer/modules/riscv64/ppp-modules | 1 + debian/installer/modules/riscv64/sata-modules | 1 + .../modules/riscv64/scsi-core-modules | 1 + debian/installer/modules/riscv64/scsi-modules | 2 + .../modules/riscv64/squashfs-modules | 1 + debian/installer/modules/riscv64/udf-modules | 1 + .../installer/modules/riscv64/uinput-modules | 1 + debian/installer/modules/riscv64/usb-modules | 1 + .../modules/riscv64/usb-storage-modules | 1 + .../installer/modules/riscv64/virtio-modules | 1 + debian/installer/modules/riscv64/zlib-modules | 1 + 41 files changed, 126 insertions(+), 1 deletion(-) create mode 100644 debian/config/riscv64/config create mode 100644 debian/config/riscv64/none/defines create mode 100644 debian/installer/modules/riscv64/ata-modules create mode 100644 debian/installer/modules/riscv64/btrfs-modules create mode 100644 debian/installer/modules/riscv64/compress-modules create mode 100644 debian/installer/modules/riscv64/crc-modules create mode 100644 debian/installer/modules/riscv64/crypto-dm-modules create mode 100644 debian/installer/modules/riscv64/crypto-modules create mode 100644 debian/installer/modules/riscv64/event-modules create mode 100644 debian/installer/modules/riscv64/ext4-modules create mode 100644 debian/installer/modules/riscv64/fat-modules create mode 100644 debian/installer/modules/riscv64/fuse-modules create mode 100644 debian/i
Bug#899118: flash-kernel: add missing arm64 boards
On Mon, May 21, 2018 at 01:01:56AM +0200, Heinrich Schuchardt wrote: > On 05/20/2018 09:15 PM, Karsten Merker wrote: > > On Sat, May 19, 2018 at 02:57:41PM +0200, Heinrich Schuchardt wrote: > > > >> Package: flash-kernel > >> Version: 3.94 > >> Severity: normal > >> Tags: patch > >> > >> Add 60 missing database entries for arm64 boards > >> supported both by U-Boot and by Linux: [...] > > many thanks for the patch. Just to make sure that we don't run > > into problems later on: do all these boards really use u-boot's > > config_distro_bootcmd.h and thereby work properly with > > bootscr.uboot-generic? > > > > When looking at the defconfigs for several of these systems, I > > see e.g. CONFIG_BOOTARGS settings that don't really match what I > > would expect for systems using config_distro_bootcmd.h. > > Three random examples: > > > > - r8a77995_draak_defconfig: > > CONFIG_BOOTARGS="console=ttySC0,115200 rw root=/dev/nfs > > nfsroot=192.168.0.1:/export/rfs ip=192.168.0.20" > > > > - ls1088ardb_sdcard_qspi_defconfig: > > CONFIG_BOOTARGS="console=ttyS0,115200 root=/dev/ram0 > > earlycon=uart8250,mmio,0x21c0500 ramdisk_size=0x300 > > default_hugepagesz=2m hugepagesz=2m hugepages=256" > > > > - hikey_defconfig: > > CONFIG_BOOTARGS="console=ttyAMA0,115200n8 root=/dev/mmcblk0p9 rw" [,,,] > For a board to be able to benefit from flash-kernel U-Boot must: > > - be capable to load and execute boot.scr > - provide the booti command > - allow the definition of the following environment variables: > - devtype, devnum, partition > - fdtfile > - kernel_addr_r > - fdt_addr_r > - ramdisk_addr_r > > In your examples above I could not find any evidence that U-Boot cannot > be configured and built to fulfill these requirements. > > For instance build > > make r8a77995_draak_defconfig > make menuconfig > CONFIG_DISTRO_DEFAULTS=y > CONFIG_BOOTARGS="console=ttySC0,115200" > CONFIG_ENV_IS_IN_MMC=y (this is a default value) > make -j6 > > Setup missing environment variables interactively and save them to MMC > and you can rely on flash-kernel for booting. > > ls1088ardb_sdcard_qspi_defconfig and hikey_defconfig select > CONFIG_DISTRO_DEFAULTS=y. CONFIG_BOOTARGS has to be reconfigured. > > This patch is only about providing flash-kernel for the boards. It is > not about providing U-Boot configured to match flash-kernel's requirements. > > I think that even for boards for which we do not provide U-Boot as a > Debian package we should allow the usage of flash-kernel without setting > up a local override for the machine database (/etc/flash-kernel/db). Hello, our policy has always been to only add an entry for an armhf/arm64 system to the flash-kernel database if this entry works out of the box with the default mainline u-boot configuration for the respective device. Users usually expect that a system boots out of the box without having to build a non-standard u-boot configuration and modify the default environment if the system is listed in the flash-kernel machine db. I'm open to arguments to the contrary, but for now I believe that is makes sense to keep this policy. Having to build u-boot with an undocumented non-standard configuration and having to modify various non-obvious parts of the default environment (which possibly includes having to find out the platform-specific kernel_addr_r, fdt_addr_r and ramdisk_addr_r values for systems where they are not already defined in the default environment) to make the system boot is an extremely high and completely unexpected hurdle for a "normal" user, and I believe that we would do a large part of our userbase a misservice by including entries that do not work out of the box. For somebody who has the knowledge to perform all the aforementioned steps on the u-boot side, it's usually easy to also add a corresponding local entry to /etc/flash-kernel/db. I'm CCing debian-arm for further opinions on the matter. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#899118: flash-kernel: add missing arm64 boards
On Sat, May 19, 2018 at 02:57:41PM +0200, Heinrich Schuchardt wrote: > Package: flash-kernel > Version: 3.94 > Severity: normal > Tags: patch > > Add 60 missing database entries for arm64 boards > supported both by U-Boot and by Linux: > > Amlogic Meson GXL (S905X) P212 Development Board > BananaPi-M64 > Freescale Layerscape 2080a QDS Board > Freescale Layerscape 2080a RDB Board > FriendlyARM NanoPi A64 > FriendlyARM NanoPi NEO 2 > FriendlyARM NanoPi NEO Plus2 > GeekBox > HiKey Development Board > HiSilicon Poplar Development Board > Khadas VIM > Libre Computer Board ALL-H3-CC H5 > Libre Technology CC > LS1012A Freedom Board > LS1012A QDS Board > LS1012A RDB Board > LS1043A RDB Board > LS1046A RDB Board > LS1088A QDS Board > LS1088A RDB Board > Marvell Armada 3720 Development Board DB-88F3720-DDR3 > Marvell Armada 7040 DB board > NVIDIA Tegra210 P2371 (P2530/P2595) reference design > NVIDIA Tegra210 P2571 reference design > Olimex A64-Olinuxino > OrangePi Win/Win Plus > OrangePi Zero Plus2 > Pine64 > Renesas Draak board based on r8a77995 > Renesas Eagle board based on r8a77970 > Renesas H3ULCB board based on r8a7795 ES2.0+ > Renesas M3ULCB board based on r8a7796 > Renesas Salvator-X board based on r8a7795 ES2.0+ > Renesas Salvator-X board based on r8a7796 > Renesas Salvator-X board based on r8a77965 > Rockchip PX5 EVB > Rockchip RK3328 EVB > SoCFPGA Stratix 10 SoCDK > UniPhier LD11 Global Board (REF_LD11_GP) > UniPhier LD11 Reference Board > UniPhier LD20 Global Board (REF_LD20_GP) > UniPhier LD20 Reference Board > UniPhier PXs3 Reference Board > Xunlong Orange Pi PC 2 > Xunlong Orange Pi Prime > ZynqMP ZC1232 RevA > ZynqMP ZC1254 RevA > ZynqMP ZC1275 RevA > ZynqMP zc1751-xm015-dc1 RevA > ZynqMP zc1751-xm016-dc2 RevA > ZynqMP zc1751-xm017-dc3 RevA > ZynqMP zc1751-xm018-dc4 > ZynqMP zc1751-xm019-dc5 RevA > ZynqMP ZCU100 RevC > ZynqMP ZCU102 Rev1.0 > ZynqMP ZCU102 RevA > ZynqMP ZCU102 RevB > ZynqMP ZCU104 RevA > ZynqMP ZCU106 RevA > ZynqMP ZCU111 RevA Hello, many thanks for the patch. Just to make sure that we don't run into problems later on: do all these boards really use u-boot's config_distro_bootcmd.h and thereby work properly with bootscr.uboot-generic? When looking at the defconfigs for several of these systems, I see e.g. CONFIG_BOOTARGS settings that don't really match what I would expect for systems using config_distro_bootcmd.h. Three random examples: - r8a77995_draak_defconfig: CONFIG_BOOTARGS="console=ttySC0,115200 rw root=/dev/nfs nfsroot=192.168.0.1:/export/rfs ip=192.168.0.20" - ls1088ardb_sdcard_qspi_defconfig: CONFIG_BOOTARGS="console=ttyS0,115200 root=/dev/ram0 earlycon=uart8250,mmio,0x21c0500 ramdisk_size=0x300 default_hugepagesz=2m hugepagesz=2m hugepages=256" - hikey_defconfig: CONFIG_BOOTARGS="console=ttyAMA0,115200n8 root=/dev/mmcblk0p9 rw" Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#890791: stretch-pu: package dpkg/1.18.25
On Sun, Apr 22, 2018 at 12:39:42AM +0200, Manuel A. Fernandez Montecelo wrote: > 2018-04-20 1:52 GMT+02:00 Guillem Jover: > > On Wed, 2018-02-28 at 18:45:49 +, Adam D. Barratt wrote: > >> On Wed, 2018-02-28 at 16:05 +0100, Manuel A. Fernandez Montecelo wrote: > >> > 2018-02-18 22:26 Guillem Jover: > >> > > I'd like to update dpkg in stretch. This includes several fixes for > >> > > documentation, regressions, misbheavior, minor security issues, and > >> > > a new arch definition so that DAK can accept packages using it. The > >> > > fixes have been in sid/buster for a while now. > >> > > >> > We depend on this version being accepted and installed in the systems > >> > where DAK lives to learn about the new architecture. After that, > >> > several other packages can add support for the architecture, without > >> > receiving an automatic reject when uploaded. > >> > > >> > It would be great if this update could enter in the next stable > >> > update, so we can make progress on that front. > > > >> We've been discussing this amongst the SRMs and are quite wary of a > >> dpkg update this close to the p-u freeze. We appreciate that the > >> changes individually seem self-contained but would like to have an > >> update of such a key package able to be tested more than is feasible in > >> the time available. > >> > >> (On a related note, in practical terms it's very unlikely that there > >> would be sufficient time to get the new strings that are introduced > >> translated.) > > > > Is there perhaps anything you are waiting for me to do here? > > > > For the strings I realized afterwards I can take the ones from sid. > > I've not yet checked how many, or if I'd really need a call for > > translation, but I'd rather do that only after I know which parts you > > are going to accept. :) But TBH, this being gettext I'm not sure it > > matters that much, as only those strings might end up not being > > translated and dpkg does not have 100% coverage for all languages > > anyway. :) [...] > So in the 2+ months since that original request, we went from > scattered packages outside the Debian infrastructure to having a port > in debian-ports infra, with >60% of packages built. > > Unfortunately, important packages like binutils, glibc or linux-* have > to stay in the separate "unreleased" suite for that reason, so the > lack of support in dak for riscv64 is causing us to do this extra > work, which would be otherwise unneeded To add some further information: having to always manually build and upload a number of otherwise perfectly working essential riscv64 packages like glibc to unreleased due to the missing dpkg/stable update doesn't only cause unnecessary additional work for the riscv64 porters but also poses practical problems for package maintainers trying to sort out portability issues as they cannot simply debootstrap a current riscv64 chroot (and therefore easily use pbuilder) because debootstrap cannot handle pulling the base system from more than one suite (unstable+unreleased in this case). Previous stretch point releases have been published every 2.5-3 months, and since the 9.4 release more than two months have passed already, so I would again like to ask the stable release managers what exactly they require to be done for the proposed dpkg update to be accepted into the 9.5 release. Guillem had asked a similar question already three weeks ago (see above), but has AFAICS received no reply to it. I would really like to avoid the dpkg/stable update necessary for proper riscv64 support being pushed back yet another release cycle because of communication issues... Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#894297: cyrus-sasl2: diff for NMU version 2.1.27~101-g0780600+dfsg-3.1
Control: tags 894297 + pending Hello, I've uploaded an NMU for cyrus-sasl2 (versioned as 2.1.27~101-g0780600+dfsg-3.1) to unstable as discussed in bug #894297. Attached is the nmudiff. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nru cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/changelog cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/changelog --- cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/changelog 2017-03-19 13:30:33.0 +0100 +++ cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/changelog 2018-04-21 18:20:16.0 +0200 @@ -1,3 +1,10 @@ +cyrus-sasl2 (2.1.27~101-g0780600+dfsg-3.1) unstable; urgency=medium + + * Non-maintainer upload with maintainer permission. + * Add support for build-profiles. (Closes: #894297) + + -- Karsten Merker <mer...@debian.org> Sat, 21 Apr 2018 18:20:16 +0200 + cyrus-sasl2 (2.1.27~101-g0780600+dfsg-3) unstable; urgency=medium [ Holger Levsen ] diff -Nru cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/control cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/control --- cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/control 2016-12-31 15:59:34.0 +0100 +++ cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/control 2018-04-21 18:20:06.0 +0200 @@ -11,17 +11,17 @@ autotools-dev, chrpath, debhelper (>= 9), - default-libmysqlclient-dev | libmysqlclient-dev, + default-libmysqlclient-dev | libmysqlclient-dev , dh-autoreconf, docbook-to-man, groff-base, - heimdal-multidev, - krb5-multidev, + heimdal-multidev , + krb5-multidev , libdb-dev, - libkrb5-dev, - libldap2-dev, + libkrb5-dev , + libldap2-dev , libpam0g-dev, - libpq-dev, + libpq-dev , libsqlite3-dev, libssl-dev, po-debconf, @@ -119,6 +119,7 @@ Depends: libsasl2-modules (= ${binary:Version}), ${misc:Depends}, ${shlibs:Depends} +Build-Profiles: Description: Cyrus SASL - pluggable authentication modules (LDAP) This is the Cyrus SASL API implementation, version 2.1. See package libsasl2-2 and RFC for more information. @@ -145,6 +146,7 @@ Depends: libsasl2-modules (= ${binary:Version}), ${misc:Depends}, ${shlibs:Depends} +Build-Profiles: Description: Cyrus SASL - pluggable authentication modules (SQL) This is the Cyrus SASL API implementation, version 2.1. See package libsasl2-2 and RFC for more information. @@ -160,6 +162,7 @@ ${misc:Depends}, ${shlibs:Depends} Conflicts: libsasl2-modules-gssapi-heimdal +Build-Profiles: Description: Cyrus SASL - pluggable authentication modules (GSSAPI) This is the Cyrus SASL API implementation, version 2.1. See package libsasl2-2 and RFC for more information. @@ -189,6 +192,7 @@ ${misc:Depends}, ${shlibs:Depends} Conflicts: libsasl2-modules-gssapi-mit +Build-Profiles: Description: Pluggable Authentication Modules for SASL (GSSAPI) This is the Cyrus SASL API implementation, version 2.1. See package libsasl2-2 and RFC for more information. diff -Nru cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/rules cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/rules --- cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/rules 2016-12-31 15:59:34.0 +0100 +++ cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/rules 2018-04-21 18:20:06.0 +0200 @@ -37,7 +37,7 @@ BDB_VERSION ?= $(shell LC_ALL=C dpkg-query -l 'libdb[45].[0-9]-dev' | grep ^ii | sed -e 's|.*\s\libdb\([45]\.[0-9]\)-dev\s.*|\1|') # SQL support may be turned off during the build, but is on by default. -ifeq (,$(findstring no-sql,$(DEB_BUILD_OPTIONS))) +ifeq (,$(findstring no-sql,$(DEB_BUILD_OPTIONS))$(findstring pkg.cyrus-sasl2.nosql,$(DEB_BUILD_PROFILES))) CONFIGURE_SQL=--enable-sql else CONFIGURE_SQL=--disable-sql @@ -45,7 +45,7 @@ endif # LDAP support may be turned off during the build, but is on by default. -ifeq (,$(findstring no-ldap,$(DEB_BUILD_OPTIONS))) +ifeq (,$(findstring no-ldap,$(DEB_BUILD_OPTIONS))$(findstring pkg.cyrus-sasl2.noldap,$(DEB_BUILD_PROFILES))) CONFIGURE_LDAP=--with-ldap CONFIGURE_LDAPDB=--enable-ldapdb else @@ -55,7 +55,7 @@ endif # GSSAPI support may be turned off during the build, but is on by default -ifeq (,$(findstring no-gssapi,$(DEB_BUILD_OPTIONS))) +ifeq (,$(findstring no-gssapi,$(DEB_BUILD_OPTIONS))$(findstring pkg.cyrus-sasl2.nogssapi,$(DEB_BUILD_PROFILES))) CONFIGURE_GSSAPI=--enable-gssapi else CONFIGURE_GSSAPI=--disable-gssapi @@ -206,7 +206,7 @@ # Alter the rpath of certain binaries and shared libraries. chrpath -d debian/tmp/usr/sbin/sasldbl
Bug#894297: Intent to NMU cyrus-sasl2 (was: [PATCH] cyrus-sasl2: please add build-profile support)
On Wed, Mar 28, 2018 at 04:53:43PM +0200, Karsten Merker wrote: [Architeture bootstrapping issues with cyrus-sasl2] > Cyrus-sasl2 is involved in a circular dependency chain whose > untangling requires building the package in multiple steps with > reduced functionality and reduced build-dependencies. To achieve > that, debian/rules currently supports passing a number of options > (no-sql, no-ldap and no-gssapi) in $(DEB_BUILD_OPTIONS). This > mechanism has the limitation that it requires manually adjusting the > build-dependency list based on the options set and makes it hard to > autobuild the package in a bootstrapping context. The use of > build-profiles (https://wiki.debian.org/BuildProfileSpec) would make > this process quite a bit easier. Attached is a patch that adds > build-profile support to the packge while keeping the original > $(DEB_BUILD_OPTIONS)-based mechanism in place. > > The build profile names follow the "extension namespace" conventions > from the build profile specification: > > DEB_BUILD_OPTIONS -> build-profile name > - > no-sql-> pkg.cyrus-sasl2.nosql > no-ldap -> pkg.cyrus-sasl2.noldap > no-gssapi -> pkg.cyrus-sasl2.nogssapi > > It would be very helpful for our bootstrapping efforts if you could > upload a version of the cyrus-sasl2 package with this patch applied to > unstable in the near future. For a standard build the patch changes > nothing, so there should be no significant risk in applying it. Hello Ondřej, unless you should object, I intend to perform an NMU of cyrus-sasl2 with the patch from bug #894297 under Low-Threshold-NMU rules on Sunday. Reasons: - Judging from https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=799864, the package changelog and the lack of activity on the pkg-cyrus-sasl2-debian-devel mailinglist, you appear to be the only remaining person in the cyrus-sasl2 packaging team. - You have marked all your packages as "Low-Threshold-NMU" at https://wiki.debian.org/LowThresholdNmu. - The patch from bug #894297 is largely noninvasive as it doesn't touch the actual cyrus-sasl2 codebase, only has any effect at all if a build-profile is explicitly enabled during building the package and uses the existing DEB_BUILD_OPTIONS-based infrastructure in debian/rules. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#896063: gcc-default-ports: cross compiler for RISC-V 32bit
On Thu, Apr 19, 2018 at 07:58:13AM +0200, Heinrich Schuchardt wrote: > Source: gcc-defaults-ports > Version: 1.176 > Severity: wishlist > > Dear Maintainer, > > we have a package to cross compile RISC-V 64-bit (gcc-riscv64-linux-gnu). > > Could you, please, also provide a similar package to cross compile RISC-V > 32-bit. > > This package is needed when developing software for 32-bit embedded RISC-V > systems. Hello, providing a gcc-riscv32-linux-gnu cross-compiler in Debian unfortunately isn't possible yet as upstream glibc currently has only support for riscv64, but not for riscv32. Making predictions about the future is obviously difficult, but from looking at the current state of development I would guess that riscv32 support probably won't make it into the glibc 2.28 release which is planned for August 1, 2018, so we are probably talking about a timeframe of "sometime in 2019". I'm CCing the debian-riscv list in case somebody there should have further insights regarding riscv32 support in glibc. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#895969: webkit2gtk: FTBFS on riscv64 due to not linking against libatomic
On Wed, Apr 18, 2018 at 11:36:16AM +0200, Alberto Garcia wrote: > On Wed, Apr 18, 2018 at 08:36:29AM +0200, Karsten Merker wrote: > > Source: webkit2gtk > > Version: 2.20.1-1 > > X-Debbugs-CC: debian-ri...@lists.debian.org > > User: debian-ri...@lists.debian.org > > Usertags: riscv64 > > > > Hello, > > > > webkit2gtk 2.20.1-1 FTBFS on the riscv64 architecture with "undefined > > reference to `__atomic_compare_exchange_1'". Full log at: > > Is there any way that I can have access to / set up a riscv64 build > environment in order to debug this problem? Hello, we don't yet have a "regular" porterbox for riscv64, but you can run a local qemu-based riscv64 virtual machine. A step-by-step description of how to set it up is available in the Debian wiki at https://wiki.debian.org/RISC-V#Setting_up_a_riscv64_virtual_machine If there should be any problems with the setup, feel free to ask me. > Or, do you have a working patch for this ? Unfortunately not. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#895934: [PATCH 1/1] flash-kernel: add Rockchip RK3288 Tinker Board
Control: tag 895934 pending On Tue, Apr 17, 2018 at 05:41:51PM +0200, Heinrich Schuchardt wrote: > Package: flash-kernel > Version: 3.94 > Severity: wishlist > Tags: patch > > Add database entry for the Tinker Board. > > Signed-off-by: Heinrich Schuchardt> --- > db/all.db | 7 +++ > 1 file changed, 7 insertions(+) > > diff --git a/db/all.db b/db/all.db > index 28d40b1..ff92e17 100644 > --- a/db/all.db > +++ b/db/all.db > @@ -412,6 +412,13 @@ Boot-Script-Path: /boot/boot.scr > U-Boot-Script-Name: bootscr.uboot-generic > Required-Packages: u-boot-tools > > +Machine: Rockchip RK3288 Tinker Board > +Kernel-Flavors: armp armmp-lpae > +DTB-Id: rk3288-tinker.dtb > +Boot-Script-Path: /boot/boot.scr > +U-Boot-Script-Name: bootscr.uboot-generic > +Required-Packages: u-boot-tools > + > Machine: Firefly-RK3399 Board > Kernel-Flavors: arm64 > DTB-Id: rk3399-firefly.dtb Hello, I've committed the patch to the flash-kernel git repository with two tiny changes: - I've fixed a typo in the kernel-flavors line (s/armp/armmp/). - I've moved the entry to another position as we (at least nominally) sort the entries by the "Machine" field and not by the "DTB-Id" field. I'll probably upload the package sometime on the upcoming weekend. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#892504: [PATCH] nspr: Please add support for the RISC-V architecture
On Mon, Apr 09, 2018 at 09:35:29PM +0200, Karsten Merker wrote: > attached is an updated packaging patch adding support for the > RISC-V architecture to nspr package version 2:4.19-1. The > original patch that had been submitted to this bug had been for > package version 2:4.18-1 and doesn't apply cleanly on version > 2:4.19-1. > > In addition to the updated packaging patch I have also attached > a "bare" patch directly against upstream. Hello, upstream has reviewed the patch and accepted it for inclusion into the upstream NSPR mercurial repository: https://hg.mozilla.org/projects/nspr/rev/f47871e2aeb1 It would be great if you could upload a new version of the Debian nspr 2:4.19 package with the patch included. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#892504: [PATCH] nspr: Please add support for the RISC-V architecture
Hello, attached is an updated packaging patch adding support for the RISC-V architecture to nspr package version 2:4.19-1. The original patch that had been submitted to this bug had been for package version 2:4.18-1 and doesn't apply cleanly on version 2:4.19-1. In addition to the updated packaging patch I have also attached a "bare" patch directly against upstream. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur nspr-4.19.orig/debian/patches/Add-type-definitions-for-the-RISC-V-architecture.patch nspr-4.19/debian/patches/Add-type-definitions-for-the-RISC-V-architecture.patch --- nspr-4.19.orig/debian/patches/Add-type-definitions-for-the-RISC-V-architecture.patch 1970-01-01 00:00:00.0 + +++ nspr-4.19/debian/patches/Add-type-definitions-for-the-RISC-V-architecture.patch 2018-04-09 18:37:58.232415922 + @@ -0,0 +1,131 @@ +From d4b3321b5eeb7976a8ca2169128a3754e3b2a8bd Mon Sep 17 00:00:00 2001 +From: Karsten Merker <mer...@debian.org> +Date: Fri, 9 Mar 2018 19:38:12 +0100 +Subject: [PATCH] Add type definitions for the RISC-V architecture. + +--- + nspr/pr/include/md/_linux.cfg | 92 +++ + nspr/pr/include/md/_linux.h | 4 ++ + 2 files changed, 96 insertions(+) + +diff --git a/nspr/pr/include/md/_linux.cfg b/nspr/pr/include/md/_linux.cfg +index b4c0ed4..afc407c 100644 +--- a/nspr/pr/include/md/_linux.cfg b/nspr/pr/include/md/_linux.cfg +@@ -1020,6 +1020,98 @@ + #define PR_BYTES_PER_WORD_LOG2 2 + #define PR_BYTES_PER_DWORD_LOG2 3 + ++#elif defined(__riscv) && (__riscv_xlen == 32) ++ ++#undef IS_BIG_ENDIAN ++#define IS_LITTLE_ENDIAN 1 ++#undef IS_64 ++ ++#define PR_BYTES_PER_BYTE 1 ++#define PR_BYTES_PER_SHORT 2 ++#define PR_BYTES_PER_INT4 ++#define PR_BYTES_PER_INT64 8 ++#define PR_BYTES_PER_LONG 4 ++#define PR_BYTES_PER_FLOAT 4 ++#define PR_BYTES_PER_DOUBLE 8 ++#define PR_BYTES_PER_WORD 4 ++#define PR_BYTES_PER_DWORD 8 ++ ++#define PR_BITS_PER_BYTE8 ++#define PR_BITS_PER_SHORT 16 ++#define PR_BITS_PER_INT 32 ++#define PR_BITS_PER_INT64 64 ++#define PR_BITS_PER_LONG32 ++#define PR_BITS_PER_FLOAT 32 ++#define PR_BITS_PER_DOUBLE 64 ++#define PR_BITS_PER_WORD32 ++ ++#define PR_BITS_PER_BYTE_LOG2 3 ++#define PR_BITS_PER_SHORT_LOG2 4 ++#define PR_BITS_PER_INT_LOG25 ++#define PR_BITS_PER_INT64_LOG2 6 ++#define PR_BITS_PER_LONG_LOG2 5 ++#define PR_BITS_PER_FLOAT_LOG2 5 ++#define PR_BITS_PER_DOUBLE_LOG2 6 ++#define PR_BITS_PER_WORD_LOG2 5 ++ ++#define PR_ALIGN_OF_SHORT 2 ++#define PR_ALIGN_OF_INT 4 ++#define PR_ALIGN_OF_LONG4 ++#define PR_ALIGN_OF_INT64 8 ++#define PR_ALIGN_OF_FLOAT 4 ++#define PR_ALIGN_OF_DOUBLE 8 ++#define PR_ALIGN_OF_POINTER 4 ++#define PR_ALIGN_OF_WORD4 ++ ++#define PR_BYTES_PER_WORD_LOG2 2 ++#define PR_BYTES_PER_DWORD_LOG2 3 ++ ++#elif defined(__riscv) && (__riscv_xlen == 64) ++ ++#undef IS_BIG_ENDIAN ++#define IS_LITTLE_ENDIAN 1 ++#define IS_64 ++ ++#define PR_BYTES_PER_BYTE 1 ++#define PR_BYTES_PER_SHORT 2 ++#define PR_BYTES_PER_INT4 ++#define PR_BYTES_PER_INT64 8 ++#define PR_BYTES_PER_LONG 8 ++#define PR_BYTES_PER_FLOAT 4 ++#define PR_BYTES_PER_DOUBLE 8 ++#define PR_BYTES_PER_WORD 8 ++#define PR_BYTES_PER_DWORD 8 ++ ++#define PR_BITS_PER_BYTE8 ++#define PR_BITS_PER_SHORT 16 ++#define PR_BITS_PER_INT 32 ++#define PR_BITS_PER_INT64 64 ++#define PR_BITS_PER_LONG64 ++#define PR_BITS_PER_FLOAT 32 ++#define PR_BITS_PER_DOUBLE 64 ++#define PR_BITS_PER_WORD64 ++ ++#define PR_BITS_PER_BYTE_LOG2 3 ++#define PR_BITS_PER_SHORT_LOG2 4 ++#define PR_BITS_PER_INT_LOG25 ++#define PR_BITS_PER_INT64_LOG2 6 ++#define PR_BITS_PER_LONG_LOG2 6 ++#define PR_BITS_PER_FLOAT_LOG2 5 ++#define PR_BITS_PER_DOUBLE_LOG2 6 ++#define PR_BITS_PER_WORD_LOG2 6 ++ ++#define PR_ALIGN_OF_SHORT 2 ++#define PR_ALIGN_OF_INT 4 ++#define PR_ALIGN_OF_LONG8 ++#define PR_ALIGN_OF_INT64 8 ++#define PR_ALIGN_OF_FLOAT 4 ++#define PR_ALIGN_OF_DOUBLE 8 ++#define PR_ALIGN_OF_POINTER 8 ++#define PR_ALIGN_OF_WORD8 ++ ++#define PR_BYTES_PER_WORD_LOG2 3 ++#define PR_BYTES_PER_DWORD_LOG2 3 ++ + #else + + #error "Unknown CPU architecture" +diff --git a/nspr/pr/include/md/_linux.h b/nspr/pr/include/md/_linux.h +index b4b298b..2370ab8 100644 +--- a/nspr/pr/include/md/_linux.h b/nspr/pr/include/md/_linux.h +@@ -57,6 +57,10 @@ + #define _PR_SI_ARCHITECTURE "m32r" + #elif defined(__or1k__) + #define _PR_SI_ARCHITECTURE "or1k" ++#elif defined(__riscv) && (__riscv_xlen == 32) ++#define _PR_SI_ARCHITECTURE "riscv32" ++#elif defined(__riscv) && (__riscv_xlen == 64) ++#define _PR_SI_ARCHITECTURE "riscv64" + #else + #error "Unknown CPU architecture" + #endif +-- +2.11.
Bug#894505: [PATCH] meson: meson-related problem when crossbuilding systemd
Package: meson Version: 0.45.1-1 Tags: patch X-Debbugs-CC: debian-ri...@lists.debian.org User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, we are in the process of bootstrapping a Debian port for the riscv64 architecture (https://wiki.debian.org/RISC-V). The systemd package - which uses meson as its build system - is part of the build-dependency chain for the essential package set, so we need to cross-build systemd to be able to complete the bootstrap process. Due to the meson issue at https://github.com/mesonbuild/meson/issues/3113 it currently isn't possible to crossbuild the systemd package in Debian. A fix for this is already in upstream meson git at https://github.com/mesonbuild/meson/commit/c4192a04fd3d46ac7a0ee81a158e7b1e3d4f06f8 but not part of the most-recent upstream meson release (0.45.1). It would be great if you could carry this one-line patch in the Debian meson package until a new upstream release including this fix is available. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#894385: [PATCH] cracklib2: Add support for a "nopython" build-profile
Package: cracklib2 Version: 2.9.2-5.1 Severity: wishlist Tags: patch X-Debbugs-CC: debian-ri...@lists.debian.org User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, we are in the process of bootstrapping a Debian port for the riscv64 architecture (https://wiki.debian.org/RISC-V). The libcrack2-dev, libcrack2 and cracklib-runtime packages are part of the build-dependency chain for the essential package set, so we need to build them to be able to complete the bootstrap process. Currently the cracklib2 source package unconditionally build-depends on python2 and python3, but those aren't available during early architecture bootstrap and are only needed for building python{,3}-cracklib, but not for cracklib-runtime, libcrack2 and libcrack2-dev. The package's debian/rules already contains support for a build without python by setting the DEB_STAGE variable to "stage1", but this mechanism has the limitation that it requires manually adjusting the build-dependency list and makes it hard to autobuild the package in a bootstrapping context. Because of this limitation, the DEB_STAGE mechanism has been deprecated; packages should now support build-profiles (https://wiki.debian.org/BuildProfileSpec). Attached is a patch that adds support for the standard "nopython" build-profile to the package while keeping the original DEB_STAGE-based mechanism in place. As the cracklib2 package is on the "Low Threshold NMU" list, I intend to do an NMU with this patch on the upcoming weekend unless you should object. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur cracklib2-2.9.2.orig/debian/control cracklib2-2.9.2/debian/control --- cracklib2-2.9.2.orig/debian/control +++ cracklib2-2.9.2/debian/control @@ -14,10 +14,10 @@ docbook-xml, dpkg-dev (>= 1.16.1~), libtool, - python-all-dev (>= 2.6.6-3~), - python-setuptools, - python3-all-dev (>= 3.1.3-2~), - python3-setuptools + python-all-dev (>= 2.6.6-3~) , + python-setuptools , + python3-all-dev (>= 3.1.3-2~) , + python3-setuptools Homepage: http://sourceforge.net/projects/cracklib Vcs-Git: https://anonscm.debian.org/git/pkg-cracklib/pkg-cracklib.git Vcs-Browser: https://anonscm.debian.org/gitweb/?p=pkg-cracklib/pkg-cracklib.git @@ -90,6 +90,7 @@ ${shlibs:Depends} Provides: ${python:Provides} Conflicts: python-crack +Build-Profiles: Description: Python bindings for password checker library cracklib2 This package provides Python bindings for cracklib. It contains a pythonic interface to cracklib's functions and some Python @@ -108,6 +109,7 @@ ${python3:Depends}, ${shlibs:Depends} Provides: ${python3:Provides} +Build-Profiles: Description: Python3 bindings for password checker library cracklib2 This package provides Python bindings for cracklib. It contains a pythonic interface to cracklib's functions and some Python diff -Nur cracklib2-2.9.2.orig/debian/rules cracklib2-2.9.2/debian/rules --- cracklib2-2.9.2.orig/debian/rules +++ cracklib2-2.9.2/debian/rules @@ -10,11 +10,13 @@ DEB_BUILD_GNU_TYPE ?= $(shell dpkg-architecture -qDEB_BUILD_GNU_TYPE) DEB_HOST_MULTIARCH ?= $(shell dpkg-architecture -qDEB_HOST_MULTIARCH) -ifneq ($(DEB_STAGE),stage1) +ifeq ($(filter stage1,$(DEB_STAGE))$(filter nopython,$(DEB_BUILD_PROFILES)),) PYVERS := $(shell pyversions -vs) PY3VERS := $(shell py3versions -vs) +DH_WITH_PARAMETERS := python2,python3,autotools_dev else NOPYTHON_OPTIONS = -Npython-cracklib -Npython3-cracklib +DH_WITH_PARAMETERS := autotools_dev endif ifeq ($(DEB_HOST_GNU_TYPE),$(DEB_BUILD_GNU_TYPE)) @@ -48,7 +50,7 @@ override_dh_auto_build: $(MAKE) -C $(CURDIR)/debian/buildtmp/base -ifneq ($(DEB_STAGE),stage1) +ifeq ($(filter stage1,$(DEB_STAGE))$(filter nopython,$(DEB_BUILD_PROFILES)),) ln -s $(CURDIR)/debian/crack.py $(CURDIR)/python; \ for i in $(PYVERS) $(PY3VERS); do \ cd $(CURDIR)/debian/buildtmp/python$$i; \ @@ -62,7 +64,7 @@ override_dh_auto_test: mkdir $(CURDIR)/debian/tmp -ifneq ($(DEB_STAGE),stage1) +ifeq ($(filter stage1,$(DEB_STAGE))$(filter nopython,$(DEB_BUILD_PROFILES)),) $(CRACKLIB_PACKER) $(CURDIR)/debian/tmp/cracklib_dict < \ $(CURDIR)/dicts/cracklib-small for i in $(PYVERS) $(PY3VERS); do \ @@ -120,7 +122,7 @@ $(CURDIR)/debian/cracklib-runtime/usr/sbin/cracklib-packer \ $(CURDIR)/debian/cracklib-runtime/usr/sbin/cracklib-unpacker -ifneq ($(DEB_STAGE),stage1) +ifeq ($(filter stage1,$(DEB_STAGE))$(filter nopython,$(DEB_BUILD_PROFILES)),) for i in $(PYVERS); do \ cd $(CURDIR)/debian/buildtmp/python$$i/python; \ python$$i setup.py install --install-layout=deb --root $(CURDIR)/debian/python-cracklib; \ @@ -139,7 +141,7 @@
Bug#894314: [PATCH] audit: Please add support for a pkg.audit.noldap build-profile
Package: audit Version: 1:2.8.2-1 Severity: wishlist X-Debbugs-CC: debian-ri...@lists.debian.org User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, we are in the process of bootstrapping a Debian port for the riscv64 architecture (https://wiki.debian.org/RISC-V). The audit package is part of the build-dependency chain for the essential package set, so we need to build it to be able to complete the bootstrap process. Unfortunately audit is involved in a circular set of build-dependencies: audit build-depends on openldap, openldap build-depends on cyrus-sasl, cyrus-sasl build-depends on pam and pam build-depends on audit, which is where the cat chases its tail :-). To break this circular dependency in a proper way, it is necessary to add a build-profile (https://wiki.debian.org/BuildProfileSpec) to audit which allows building the package without support for openldap. AIUI, the only binary package built from the audit source that actually needs openldap is audispd-plugins, in particular the zos-remote plugin. Attached is a patch that adds support for a pkg.audit.noldap build-profile that does two things when enabled: - disable the build-dependency on libldap2-dev - disable building the audispd-plugins binary package It would be great if you could upload a version of audit with this patch included to unstable soon as we depend on it for continuing with our bootstrapping efforts. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur audit-2.8.2.orig/debian/control audit-2.8.2.new/debian/control --- audit-2.8.2.orig/debian/control +++ audit-2.8.2.new/debian/control @@ -10,7 +10,7 @@ # audit sources embed their own patched version of libev # libev-dev, libkrb5-dev, - libldap2-dev, + libldap2-dev , libprelude-dev, libwrap0-dev, python-all-dev:any (>= 2.6.6-3~) , @@ -138,6 +138,7 @@ Section: admin Architecture: linux-any Depends: auditd, ${misc:Depends}, ${shlibs:Depends} +Build-Profiles: Description: Plugins for the audit event dispatcher The audispd-plugins package provides plugins for the real-time interface to the audit system, audispd. These plugins can do things diff -Nur audit-2.8.2.orig/debian/rules audit-2.8.2.new/debian/rules --- audit-2.8.2.orig/debian/rules 2017-12-18 08:13:02.0 + +++ audit-2.8.2.new/debian/rules 2018-03-28 17:38:29.354936110 + @@ -30,6 +30,10 @@ EXTRA_ARCH_TABLE := --with-hppa endif +ifneq ($(filter pkg.audit.noldap,$(DEB_BUILD_PROFILES)),) + CONFIGURE_FLAGS += --disable-zos-remote +endif + %: dh $@ --builddirectory=debian/build --buildsystem=autoconf $(DH_ADDONS)
Bug#894297: [PATCH] cyrus-sasl2: please add build-profile support
Package: cyrus-sasl2 Version: 2.1.27~101-g0780600+dfsg-3 Severity: wishlist Tags: patch X-Debbugs-CC: debian-ri...@lists.debian.org User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, we are in the process of bootstrapping a Debian port for the riscv64 architecture (https://wiki.debian.org/RISC-V). The cyrus-sasl2 package is part of the build-dependency chain for the essential package set, so we need to build it to be able to complete the bootstrap process. Cyrus-sasl2 is involved in a circular dependency chain whose untangling requires building the package in multiple steps with reduced functionality and reduced build-dependencies. To achieve that, debian/rules currently supports passing a number of options (no-sql, no-ldap and no-gssapi) in $(DEB_BUILD_OPTIONS). This mechanism has the limitation that it requires manually adjusting the build-dependency list based on the options set and makes it hard to autobuild the package in a bootstrapping context. The use of build-profiles (https://wiki.debian.org/BuildProfileSpec) would make this process quite a bit easier. Attached is a patch that adds build-profile support to the packge while keeping the original $(DEB_BUILD_OPTIONS)-based mechanism in place. The build profile names follow the "extension namespace" conventions from the build profile specification: DEB_BUILD_OPTIONS -> build-profile name - no-sql-> pkg.cyrus-sasl2.nosql no-ldap -> pkg.cyrus-sasl2.noldap no-gssapi -> pkg.cyrus-sasl2.nogssapi It would be very helpful for our bootstrapping efforts if you could upload a version of the cyrus-sasl2 package with this patch applied to unstable in the near future. For a standard build the patch changes nothing, so there should be no significant risk in applying it. JFTR one thing that I have found while testing the patch: the three options/profiles are in practice not completely independent from each other. The main purpose of the no-gssapi option is to remove the dependency on kerberos, but if the package is built with SQL support (i.e. pkg.cyrus-sasl2.nogssapi is set but pkg.cyrus-sasl2.nosql is not), libpq5 by default pulls in krb5 nonetheless. The same limitation applies to the existing mechanism, so this isn't a regression compared to what we have now. It definitely makes sense to have separate options/profiles for no-gssapi and no-sql though, as postgresql supports a stage1 build without krb5 for bootstrapping purposes and in this case they are really independent from each other. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur cyrus-sasl2-2.1.27~101-g0780600+dfsg.orig/debian/control cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/control --- cyrus-sasl2-2.1.27~101-g0780600+dfsg.orig/debian/control +++ cyrus-sasl2-2.1.27~101-g0780600+dfsg/debian/control @@ -11,17 +11,17 @@ autotools-dev, chrpath, debhelper (>= 9), - default-libmysqlclient-dev | libmysqlclient-dev, + default-libmysqlclient-dev | libmysqlclient-dev , dh-autoreconf, docbook-to-man, groff-base, - heimdal-multidev, - krb5-multidev, + heimdal-multidev , + krb5-multidev , libdb-dev, - libkrb5-dev, - libldap2-dev, + libkrb5-dev , + libldap2-dev , libpam0g-dev, - libpq-dev, + libpq-dev , libsqlite3-dev, libssl-dev, po-debconf, @@ -119,6 +119,7 @@ Depends: libsasl2-modules (= ${binary:Version}), ${misc:Depends}, ${shlibs:Depends} +Build-Profiles: Description: Cyrus SASL - pluggable authentication modules (LDAP) This is the Cyrus SASL API implementation, version 2.1. See package libsasl2-2 and RFC for more information. @@ -145,6 +146,7 @@ Depends: libsasl2-modules (= ${binary:Version}), ${misc:Depends}, ${shlibs:Depends} +Build-Profiles: Description: Cyrus SASL - pluggable authentication modules (SQL) This is the Cyrus SASL API implementation, version 2.1. See package libsasl2-2 and RFC for more information. @@ -160,6 +162,7 @@ ${misc:Depends}, ${shlibs:Depends} Conflicts: libsasl2-modules-gssapi-heimdal +Build-Profiles: Description: Cyrus SASL - pluggable authentication modules (GSSAPI) This is the Cyrus SASL API implementation, version 2.1. See package libsasl2-2 and RFC for more information. @@ -189,6 +192,7 @@ ${misc:Depends}, ${shlibs:Depends} Conflicts: libsasl2-modules-gssapi-mit +Build-Profiles: Description: Pluggable Authentication Modules for SASL (GSSAPI) This is the
Bug#891341: libatomic-ops: [PATCH] Please add support for new architecture "riscv64" (RISC-V 64 bits little-endian)
On Sat, Feb 24, 2018 at 06:09:30PM +0100, Manuel A. Fernandez Montecelo wrote: > Source: libatomic-ops > Version: 7.6.2-1 > Severity: normal > Tags: patch upstream fixed-upstream > User: debian-ri...@lists.debian.org > Usertags: riscv64 > Forwarded: > https://github.com/ivmai/libatomic_ops/commit/27ca1b0f306cb99f14f863b22d490bb30c778300 > > Hello, > > We need support in this package for RISC-V, to bootstrap the riscv64 > architecture. > > A patch has been submitted upstream a few days ago, so it would be great if > you > could include it as a patch and release a new version for unstable. > > If we can help by NMUing the package or anything else, please let me/us know. > > Thanks and cheers. > -- > Manuel A. Fernandez MonteceloHello Ian, as one month has passed since the submission above I wanted to kindly ask you about the state of affairs regarding this bug. Libatomic-ops is a hard requirement for building the essential package set and a lot of packages transitively build-depend on it, so the Debian riscv64 port cannot move forward without having RISC-V support in the libatomic-ops package. Attached is a patch against the Debian packaging that adds four patches cherry-picked from upstream as quilt patches: https://github.com/ivmai/libatomic_ops/commit/27ca1b0f306cb99f14f863b22d490bb30c778300 https://github.com/ivmai/libatomic_ops/commit/84ef64ebbf3df35756d083d08620b167547d6860 https://github.com/ivmai/libatomic_ops/commit/4f68118ead90f5a2940cea6baa28d24ab2fb9f74 https://github.com/ivmai/libatomic_ops/commit/393d7a7ff54565230f44e59a9c73addc9e627f56 It would be great if you could upload a new version of libatomic-ops to unstable with these patches applied, so that we can move forward with our port. Kind regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur libatomic-ops-7.6.2.orig/debian/patches/riscv-0001-Add-RISC-V-support.patch libatomic-ops-7.6.2/debian/patches/riscv-0001-Add-RISC-V-support.patch --- libatomic-ops-7.6.2.orig/debian/patches/riscv-0001-Add-RISC-V-support.patch +++ libatomic-ops-7.6.2/debian/patches/riscv-0001-Add-RISC-V-support.patch @@ -0,0 +1,65 @@ +From 235b22468d0cdc0718feed41067a5ceeb7079f98 Mon Sep 17 00:00:00 2001 +From: Shea Levy +Date: Sun, 18 Feb 2018 00:47:44 -0500 +Subject: [PATCH 1/4] Add RISC-V support +Origin: https://github.com/ivmai/libatomic_ops/commit/27ca1b0f306cb99f14f863b22d490bb30c778300 + +Issue #31 (libatomic_ops). + +* src/Makefile.am (nobase_private_HEADERS): Add riscv.h entry. +* src/atomic_ops.h [__riscv]: Include riscv.h file. +* src/atomic_ops/sysdeps/gcc/riscv.h: New file (just include generic.h). +--- + src/Makefile.am| 1 + + src/atomic_ops.h | 3 +++ + src/atomic_ops/sysdeps/gcc/riscv.h | 12 + 3 files changed, 16 insertions(+) + create mode 100644 src/atomic_ops/sysdeps/gcc/riscv.h + +diff --git a/src/Makefile.am b/src/Makefile.am +index dadc932..084dd3c 100644 +--- a/src/Makefile.am b/src/Makefile.am +@@ -92,6 +92,7 @@ nobase_private_HEADERS = atomic_ops/ao_version.h \ + atomic_ops/sysdeps/gcc/mips.h \ + atomic_ops/sysdeps/gcc/nios2.h \ + atomic_ops/sysdeps/gcc/powerpc.h \ ++ atomic_ops/sysdeps/gcc/riscv.h \ + atomic_ops/sysdeps/gcc/s390.h \ + atomic_ops/sysdeps/gcc/sh.h \ + atomic_ops/sysdeps/gcc/sparc.h \ +diff --git a/src/atomic_ops.h b/src/atomic_ops.h +index 187e1f9..22e516f 100644 +--- a/src/atomic_ops.h b/src/atomic_ops.h +@@ -352,6 +352,9 @@ + # if defined(__tile__) + # include "atomic_ops/sysdeps/gcc/tile.h" + # endif ++# if defined(__riscv) ++# include "atomic_ops/sysdeps/gcc/riscv.h" ++# endif + #endif /* __GNUC__ && !AO_USE_PTHREAD_DEFS */ + + #if (defined(__IBMC__) || defined(__IBMCPP__)) && !defined(__GNUC__) \ +diff --git a/src/atomic_ops/sysdeps/gcc/riscv.h b/src/atomic_ops/sysdeps/gcc/riscv.h +new file mode 100644 +index 000..412c395 +--- /dev/null b/src/atomic_ops/sysdeps/gcc/riscv.h +@@ -0,0 +1,12 @@ ++/* ++ * THIS MATERIAL IS PROVIDED AS IS, WITH ABSOLUTELY NO WARRANTY EXPRESSED ++ * OR IMPLIED. ANY USE IS AT YOUR OWN RISK. ++ * ++ * Permission is hereby granted to use or copy this program ++ * for any purpose, provided the above notices are retained on all copies. ++ * Permission to modify the code and to distribute modified code is granted, ++ * provided the above notices are retained, and a notice that the code was ++ * modified is included with the above copyright notice. ++ */ ++ ++#include "generic.h" +-- +2.11.0 + diff -Nur libatomic-ops-7.6.2.orig/debian/patches/riscv-0002-Update-AUTHORS-file-add-Shea-Levy.patch libatomic-ops-7.6.2/debian/patches/riscv-0002-Update-AUTHORS-file-add-Shea-Levy.patch ---
Bug#893300: cdebconf: Adding support for a pkg.cdebconf.nogtk build-profile
On Mon, Mar 26, 2018 at 12:47:18AM +0100, Colin Watson wrote: > On Sat, Mar 17, 2018 at 09:09:11PM +0100, Karsten Merker wrote: > > I would like to add support for a "pkg.cdebconf.nogtk" build-profile > > to cdebconf. Background for that is that cdebconf (in particular > > libdebconfclient0) is needed rather early in the process of > > bootstrapping a new Debian architecture, but getting it built during > > early architecture bootstrap is difficult due to its build-dependency > > on gtk+cairo, which pulls in an enormous list of transitive > > build-dependencies that are effectively impossible to fullfill in a > > bootstrap scenario. > > This approach and patch looks good to me. I'm OK with you committing > and uploading it, modulo the comments below. > > > diff --git a/debian/rules b/debian/rules > > index b2b35f4d..8b85a7af 100755 > > --- a/debian/rules > > +++ b/debian/rules > > @@ -21,6 +21,11 @@ LIBDEBCONF=libdebconfclient0 > > DEB_FRONTENDS=passthrough text newt gtk > > UDEB_FRONTENDS=passthrough text newt gtk > > > > +ifneq ($(filter pkg.cdebconf.nogtk,$(DEB_BUILD_PROFILES)),) > > +DEB_FRONTENDS:=$(filter-out gtk,$(DEB_FRONTENDS)) > > +UDEB_FRONTENDS:=$(filter-out gtk,$(UDEB_FRONTENDS)) > > +endif > > I think this would be clearer reversed, i.e.: > > DEB_FRONTENDS=passthrough text newt > UDEB_FRONTENDS=passthrough text newt > > ifeq ($(filter pkg.cdebconf.nogtk,$(DEB_BUILD_PROFILES)),) > DEB_FRONTENDS+=gtk > UDEB_FRONTENDS+=gtk > endif That's probably a matter of taste :-). I found it clearer to have the primary DEB_FRONTENDS and UDEB_FRONTENDS assignments to always represent the default case when no build-profiles are enabled and only modify them in the "non-standard" case of having a build-profile set. If you have a strong preference for the "additive" version instead of the "subtractive" version, please let me know and I'll change that. > > +ifneq ($(filter pkg.cdebconf.nogtk,$(DEB_BUILD_PROFILES)),) > > + dh_install -plibdebconfclient0-dev > > src/modules/frontend/gtk/cdebconf_gtk.h usr/include/cdebconf/ > > +endif > > I think I may understand what this is doing now after some confusion, > but it's pretty opaque and definitely needs at least a comment. I think > you're trying to keep the package contents identical regardless of build > profile, since the main build system won't handle it in this case due to > the change in *_FRONTENDS? Exactly. I'll add a comment explaining that. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#891803: cracklib2: diff for NMU version 2.9.2-5.1
Control: tags 891803 + patch Control: tags 891803 + pending Dear maintainer, I've prepared an NMU for cracklib2 (versioned as 2.9.2-5.1) and will upload it now. Attached is the nmudiff. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nru cracklib2-2.9.2/debian/changelog cracklib2-2.9.2/debian/changelog --- cracklib2-2.9.2/debian/changelog 2017-05-27 11:41:18.0 +0200 +++ cracklib2-2.9.2/debian/changelog 2018-03-21 22:39:38.0 +0100 @@ -1,3 +1,11 @@ +cracklib2 (2.9.2-5.1) unstable; urgency=medium + + * Non-maintainer upload. + * Remove the ":native" annotation from the "cracklib-runtime " +build-dependency. (Closes: #891803) + + -- Karsten Merker <mer...@debian.org> Wed, 21 Mar 2018 22:39:38 +0100 + cracklib2 (2.9.2-5) unstable; urgency=medium * Add Breaks: cracklib-runtime (<< 2.9.2-4) to libcrack2 to configure diff -Nru cracklib2-2.9.2/debian/control cracklib2-2.9.2/debian/control --- cracklib2-2.9.2/debian/control 2017-05-27 11:06:18.0 +0200 +++ cracklib2-2.9.2/debian/control 2018-03-21 22:39:20.0 +0100 @@ -8,7 +8,7 @@ automake (>= 1.10), autotools-dev, chrpath, - cracklib-runtime:native , + cracklib-runtime , debhelper (>= 9), docbook-utils, docbook-xml,
Bug#890791: stretch-pu: package dpkg/1.18.25
On Wed, Feb 28, 2018 at 06:45:49PM +, Adam D. Barratt wrote: > On Wed, 2018-02-28 at 16:05 +0100, Manuel A. Fernandez Montecelo wrote: > [..] > > 2018-02-18 22:26 Guillem Jover: > [...] > > > I'd like to update dpkg in stretch. This includes several fixes for > > > documentation, regressions, misbheavior, minor security issues, and > > > a new arch definition so that DAK can accept packages using it. The > > > fixes have been in sid/buster for a while now. > > > > We depend on this version being accepted and installed in the systems > > where DAK lives to learn about the new architecture. After that, > > several other packages can add support for the architecture, without > > receiving an automatic reject when uploaded. > > > > It would be great if this update could enter in the next stable > > update, so we can make progress on that front. > > We've been discussing this amongst the SRMs and are quite wary of a > dpkg update this close to the p-u freeze. We appreciate that the > changes individually seem self-contained but would like to have an > update of such a key package able to be tested more than is feasible in > the time available. > > (On a related note, in practical terms it's very unlikely that there > would be sufficient time to get the new strings that are introduced > translated.) > > We understand that this is inconvenient for the riscv porters, so are > exploring whether it would be possible to have the dak support made > available via p-u after the upcoming point release. Hello, I wanted to kindly ask whether there are any news on this topic and whether there is anything that the RISC-V porters can do to help. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#721358: bug#28574: coreutils: use dummy man when cross build
Control: user debian-ri...@lists.debian.org Control: usertag -1 + riscv64 [coreutils not cross-buildable due to help2man issues] On Sat, Sep 30, 2017 at 08:56:53PM -0700, Pádraig Brady wrote: > On 23/09/17 20:18, Peter Bohning wrote: > > Cross compiling for aarch64 from amd64 > > > > Is there no way to disable the man pages in configure options? Thanks. > > > > https://lists.gnu.org/archive/html/coreutils/2014-11/msg0.html > > I've seen this issue in various places now (CC'd). > > We should probably reinstate distribution of man pages > since these generated pages are so similar on various systems. > Also given the amount of old ChangeLogs removed recently > I don't think the addition of 273K (uncompressed) is prohibitive. > > The attached was tested in these modes: > git repo build > tarball build with/without help2man > tarball vpath build with/without help2man > make distcheck > make install (and ensure only enabled program man pages installed) > > I've not applied this yet, but hope to soon after a bit more testing. > > cheers, > Pádraig Hello, we are currently in the process of bootstrapping a new Debian architecture (riscv64, https://wiki.debian.org/RISC-V) and therefore affected by this problem. The underlying issue has been solved in upstream coreutils 8.29, so I would like to ask whether it would perhaps be possible for you to update coreutils in unstable to version 8.29. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#891803: [PATCH] cracklib2: please remove the ":native" annotation from the cracklib-runtime build-dependency for cross-builds
On Thu, Mar 01, 2018 at 12:07:39AM +0100, Karsten Merker wrote: > Source: cracklib2 > Version: 2.9.2-5 > Severity: normal > Tags: patch > X-Debbugs-CC: debian-ri...@lists.debian.org > User: debian-ri...@lists.debian.org > Usertags: riscv64 > > Hello, > > the cracklib2 build-dependency list in debian/control currently > contains the following entry: > > Build-Depends: cracklib-runtime:native > > The cracklib-runtime package is (correctly) marked as "Multi-Arch: > foreign". Because of this the ":native" annotation in the > cracklib2 build-depends needs to be removed as it creates an > unresolvable cross-dependency due to the fact that dpkg doesn't > consider M-A:foreign packages as valid candidates for fulfilling a > dependency with a ":native" annotation. > > For more information regarding this please cf. > https://wiki.debian.org/Multiarch/MissingRationale#Why_do_M-A:foreign_packages_not_satisfy_:native_qualified_dependencies_in_build-dependencies.3F > > Attached is a corresponding patch against the current > pkg-cracklib git head. Hello, I've just seen that the package is on the lowNMU list, so unless you should object, I intend to do an NMU within the next days. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#893300: cdebconf: Adding support for a pkg.cdebconf.nogtk build-profile
Package: cdebconf Version: 0.241 Priority: wishlist X-Debbugs-CC: Cyril Brulebois <k...@debian.org>, Christian Perrier <bubu...@debian.org>, Regis Boudin <re...@debian.org>, Colin Watson <cjwat...@debian.org>, debian-ri...@lists.debian.org User: debian-ri...@lists.debian.org Usertags: riscv64 [CCing the uploaders for cdebconf] Hello, I would like to add support for a "pkg.cdebconf.nogtk" build-profile to cdebconf. Background for that is that cdebconf (in particular libdebconfclient0) is needed rather early in the process of bootstrapping a new Debian architecture, but getting it built during early architecture bootstrap is difficult due to its build-dependency on gtk+cairo, which pulls in an enormous list of transitive build-dependencies that are effectively impossible to fullfill in a bootstrap scenario. AIUI, the only binary packages built from the cdebconf source package that actually need gtk+cairo are cdebconf-gtk and cdebconf-gtk-udeb, and these aren't required during an architecture bootstrap, so the approach is to add a build-profile that does two things when set: - disable building of these two binary packages - remove the gtk+cairo build-dependency Attached is a patch that implements that. As nothing changes when the build-profile isn't explicitly activiated, this should be a low-risc modification, but as I normally don't work on cdebconf, I would like to gather feedback from the regular uploaders (in CC) whether you see some reason to object to this change. The debdiff between a standard build and a build with the build-profile set is clean; the only difference with the build-profile enabled is that the gtk-related binary packages aren't built. If the patch is ok for you, I'll apply it to cdebconf git and upload a new version. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. >From acd6f9d3065137727e7c372c306cda67adfff9b9 Mon Sep 17 00:00:00 2001 From: Karsten Merker <mer...@debian.org> Date: Thu, 15 Mar 2018 20:10:33 + Subject: [PATCH] Add a pkg.cdebconf.nogtk build-profile. When the pkg.cdebconf.nogtk profile is set, the build-dependency on gtk and cairo gets removed and the cdebconf-gtk and cdebconf-gtk-udeb binary packages (which are the only ones actually depending on gtk and cairo) aren't built. This is important when bootstrapping a new architecture as cdebconf is required for building the essential package set and a dependency on gtk and cairo pulls in an enormous list of transitive build-dependencies that are effectively impossible to fullfill in a bootstrap scenario. --- debian/control | 6 -- debian/rules | 8 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/debian/control b/debian/control index 1a194849..b87c0861 100644 --- a/debian/control +++ b/debian/control @@ -9,8 +9,8 @@ Build-Depends: libtextwrap-dev (>= 0.1-5), libdebian-installer4-dev (>= 0.41) | libdebian-installer-dev, libglib2.0-dev (>= 2.31), - libgtk2.0-dev (>= 2.24), - libcairo2-dev (>= 1.8.10-3), + libgtk2.0-dev (>= 2.24) , + libcairo2-dev (>= 1.8.10-3) , libselinux1-dev (>= 2.3) [linux-any] | libselinux-dev [linux-any], dh-autoreconf, dh-exec, @@ -42,6 +42,7 @@ Section: admin Depends: cdebconf, ${shlibs:Depends}, ${misc:Depends} Replaces: cdebconf (<< 0.144) Priority: extra +Build-Profiles: Description: Gtk+ frontend for Debian Configuration Management System Debconf is a configuration management system for Debian packages. It is used by some packages to prompt you for information before they are @@ -151,6 +152,7 @@ Architecture: any Section: debian-installer Depends: cdebconf-udeb, ${shlibs:Depends}, ${misc:Depends}, rootskel-gtk [!s390 !s390x] Package-Type: udeb +Build-Profiles: Description: Gtk+ frontend for Debian Configuration Management System Debconf is a configuration management system for Debian packages. It is used by some packages to prompt you for information before they are diff --git a/debian/rules b/debian/rules index b2b35f4d..8b85a7af 100755 --- a/debian/rules +++ b/debian/rules @@ -21,6 +21,11 @@ LIBDEBCONF=libdebconfclient0 DEB_FRONTENDS=passthrough text newt gtk UDEB_FRONTENDS=passthrough text newt gtk +ifneq ($(filter pkg.cdebconf.nogtk,$(DEB_BUILD_PROFILES)),) +DEB_FRONTENDS:=$(filter-out gtk,$(DEB_FRONTENDS)) +UDEB_FRONTENDS:=$(filter-out gtk,$(UDEB_FRONTENDS)) +endif + ifeq ($(DEB_HOST_ARCH_OS),linux) SELINUXFLAG=--enable-selinux else @@ -128,6 +133,9 @@ binary-arch: install-arch dh_installdocs -s dh_installdebconf -s dh_installdirs -s +ifneq ($(filter pkg.cdebconf.nogtk,$(DEB_BUILD_PROFILES)),) + dh_install -plibdebconfclient0-dev src/modules/frontend/gtk/cdebconf_gtk.h usr/include/cdebconf/ +endif dh_lintian -s dh_strip -s dh_compress -s -- 2.11.0
Bug#892935: libdebian-installer: Please add support for "nodoc" build options and profile
Control: tags 892935 pending On Wed, Mar 14, 2018 at 06:29:24PM +0100, Manuel A. Fernandez Montecelo wrote: > Source: libdebian-installer > Version: 0.114 > Severity: wishlist > Tags: patch > User: debian-ri...@lists.debian.org > Usertags: riscv64 > > Hi, > > We need changes in this package to help to bootstrap the riscv64 architecture, > in particular it will help to skip using doxygen (in a clean way) for the time > being. > > It will be available at some point in the future, but it's complicate to have > it > in the initial stages, due to the build-depends needed (particularly > yui-compressor and ruby-compass at this moment). This is probably the case > for > any other architecture being bootstrapped now or in the near future. > > I am attaching a patch that makes possible to compile the package by using > profiles (or options) "nodoc", avoiding the dependency in d/control (with > profile) and avoiding those files. The profile helps to avoid the > build-dependency entirel, not needing to force to build with missing > dependencies. > > Besides that, "nodoc" is a standard build option that would be nice to have in > any case. > > Please consider applying the patch (that I will upload in minutes, to have a > bug > number) or providing some alternative solution to the same effect. Looks good to me and local testbuilds haven't shown any problems. The patch has been committed to the libdebian-installer git repository; the CI build on jenkins was successful. I intend to upload the package with the patch included on the weekend unless somebody @debian-boot should voice objections. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#892593: [PATCH] libverto: FTCBFS / Please add a pkg.libverto.noglib build profile
Source: libverto Version: 0.2.4-2.1 Severity: wishlist X-Debbugs-CC: debian-ri...@lists.debian.org Tags: patch User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, we are in the process of bootstrapping a Debian port for the riscv64 architecture (https://wiki.debian.org/RISC-V). The binary packages libverto-dev and libverto-libev1 are part of the build-dependency chain for the essential package set, so we need to cross-build them for riscv64 to be able to complete the bootstrap process. Unfortunately it is currently impossible to cross-build libverto as-is due to the following situation: The package curently has a "hard" build-dependency on libglib2.0-dev, which is only needed for building libverto-glib1, but not for the other binary packages built from the libverto source package. For cross-building libverto-glib1, one would need to install libglib2.0-dev for the host architecture, but that isn't possible due to components of glib2.0 pulled in by that not being multiarch-coinstallable with their build-arch equivalents and debhelper transitively depending on the same components for the build-arch. The solution for this issue is adding a pkg.libverto.noglib build-profile to the package, which disables building the libverto-glib1 binary package and the corresponding build-dependency when set. Attached is a patch implementing that. It would be great if you could upload a new version of libverto with the patch applied to unstable soon as we depend on this for completing our bootstrapping efforts. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur libverto-0.2.4/debian/control libverto-0.2.4.patched/debian/control --- libverto-0.2.4/debian/control 2015-10-29 13:00:37.0 + +++ libverto-0.2.4.patched/debian/control 2018-03-11 08:09:28.721190904 + @@ -1,7 +1,7 @@ Source: libverto Priority: optional Maintainer: Sam Hartman-Build-Depends: debhelper (>= 9), libev-dev, libglib2.0-dev, dh-autoreconf +Build-Depends: debhelper (>= 9), libev-dev, libglib2.0-dev , dh-autoreconf Build-Conflicts: libevent-dev, libtevent-dev Standards-Version: 3.9.4 Section: libs @@ -58,6 +58,7 @@ PRe-Depends: ${misc:Pre-Depends} Depends: ${shlibs:Depends}, ${misc:Depends} Multi-Arch: same +Build-Profiles: Description: Event loop abstraction for Libraries - glib Libverto exists to isolate libraries from the particular event loop chosen by an application. Libverto provides an asynchronous diff -Nur libverto-0.2.4/debian/rules libverto-0.2.4.patched/debian/rules --- libverto-0.2.4/debian/rules 2015-10-29 12:14:17.0 + +++ libverto-0.2.4.patched/debian/rules 2018-03-11 08:09:56.128680957 + @@ -9,9 +9,13 @@ # Uncomment this to turn on verbose mode. #export DH_VERBOSE=1 +ifeq (,$(filter libverto-glib1, $(shell dh_listpackages))) + CONFOPTS += --without-glib +endif + %: dh $@ --with autoreconf override_dh_auto_configure: chmod a+x configure - dh_auto_configure -- --libdir=/usr/lib/$(shell dpkg-architecture -qDEB_HOST_MULTIARCH) + dh_auto_configure -- $(CONFOPTS) --libdir=/usr/lib/$(shell dpkg-architecture -qDEB_HOST_MULTIARCH)
Bug#892566: [PATCH] make-dfsg: Please add support for a "noguile" build-profile.
Source: make-dfsg Version: 4.2.1-1 Severity: wishlist X-Debbugs-CC: debian-ri...@lists.debian.org User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, the make-dfsg source package currently has a "hard" build-dependency on guile-2.0-dev, which is only used for building the "make-guile" binary package. During bootstrapping a new architecture one needs to build "make" but not "make-guile", and being able to do so easily without requiring guile and its build-dependencies for the new architecture would be really helpful. Attached is a trivial patch that adds a "noguile" build profile to the make-dfsg package. It just modifies debian/control as debian/rules already has all necessary infrastructure in place. It would be great if you could upload a new version of make-dfsg with this patch and the glibc-2.27 build fix from https://bugs.debian.org/891365 to unstable. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. >From aafe1a5ca4f95a2d12e13fdb28c4f3ba63bd7c3c Mon Sep 17 00:00:00 2001 From: Karsten Merker <mer...@debian.org> Date: Sat, 10 Mar 2018 16:59:30 +0100 Subject: [PATCH] Add a "noguile" build-profile. The "noguile" profile results in dropping the build-dependency on guile-2.0-dev and disabling the creation of the "make-guile" binary package. This eases (cross-)building the make package when bootstrapping a new architecture. --- debian/control | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/debian/control b/debian/control index 04aad43..a92c92a 100644 --- a/debian/control +++ b/debian/control @@ -8,7 +8,8 @@ Standards-Version: 4.1.3 Homepage: http://www.gnu.org/software/make/ Build-Depends: gettext, po-debconf, debhelper (>= 9.0.0), dh-autoreconf, autoconf, automake | automaken, autopoint, file, pkg-config, - guile-2.0-dev, procps, libbsd-resource-perl + procps, libbsd-resource-perl , + guile-2.0-dev Package: make Suggests: make-doc @@ -37,6 +38,7 @@ Provides: make Replaces: make Architecture: any Multi-Arch: allowed +Build-Profiles: Description: utility for directing compilation with guile support GNU Make is a utility which controls the generation of executables and other target files of a program from the program's source -- 2.11.0
Bug#892217: [PATCH] libffi: Please add support for the riscv64 architecture
Source: libffi Version: 3.3~20180131 Severity: wishlist X-Debbugs-CC: debian-ri...@lists.debian.org Tags: patch User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, we are in the process of bootstrapping a Debian port for the riscv64 architecture (https://wiki.debian.org/RISC-V). The libffi package is part of the build-dependency chain for the essential package set, so we need to build it for riscv64 to be able to complete the bootstrap process. Current upstream libffi doesn't yet have support for the RISC-V architecture, but a pull request to add it has been submitted: https://github.com/libffi/libffi/pull/281/ respectively https://patch-diff.githubusercontent.com/raw/libffi/libffi/pull/281.patch The pull request has been authored by one of the Fedora riscv64 porters and the Fedora riscv64 port already uses this patchset. It would be great if you could add the patchset to the Debian libffi package as well. The RISC-V support provided by the patchset is not all-encompassing (please cf. the last comment from Stefan O'Rear in the pull request), but it provides the same amount of features that libffi has for a number of other Debian-supported architectures, so that shouldn't be a problem. The symbols file would need to be adjusted, though: dh_makeshlibs -plibffi7 --add-udeb=libffi7-udeb dpkg-gensymbols: warning: some symbols or patterns disappeared in the symbols file: see diff output below dpkg-gensymbols: warning: debian/libffi7/DEBIAN/symbols doesn't match completely debian/libffi7.symbols --- debian/libffi7.symbols (libffi7_3.3~20180131-2_riscv64) +++ dpkg-gensymbols50ytax 2018-03-06 21:11:57.533480937 + @@ -2,5 +2,5 @@ (symver)LIBFFI_BASE_7.0 3.3~20170512 (symver)LIBFFI_BASE_7.1 3.3~20170512 (symver)LIBFFI_CLOSURE_7.0 3.3~20170512 - (symver|arch=!hppa !ia64 !m68k !mips !mipsel !mips64el !powerpc !powerpcspe !ppc64 !ppc64el !sh4)LIBFFI_COMPLEX_7.0 3.3~20170512 - (symver|arch=!hppa !ia64 !m68k !sh4)LIBFFI_GO_CLOSURE_7.0 3.3~20170512 +#MISSING: 3.3~20180131-2# (symver|arch=!hppa !ia64 !m68k !mips !mipsel !mips64el !powerpc !powerpcspe !ppc64 !ppc64el !sh4)LIBFFI_COMPLEX_7.0 3.3~20170512 +#MISSING: 3.3~20180131-2# (symver|arch=!hppa !ia64 !m68k !sh4)LIBFFI_GO_CLOSURE_7.0 3.3~20170512 Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#891663: [PATCH] libgpg-error: Add support for the riscv64 architecture
Source: libgpg-error Version: 1.27-6 X-Debbugs-CC: debian-ri...@lists.debian.org Tags: patch Severity: wishlist User: debian-ri...@lists.debian.org Usertags: riscv64 Hello, we are in the process of bootstrapping a Debian port for the riscv64 architecture (https://wiki.debian.org/RISC-V). The libgpg-error package is part of the build-dependency chain for the essential package set, so we need to build it for riscv64 to be able to complete the bootstrap process. AIUI, libgpg-error requires a per-architecture header file "src/syscfg/lock-obj-pub..h", which upstream doesn't provide for riscv64. Based on the information in the header files for the other architectures, I have created such a file for riscv64 by cross-building and executing (in qemu) the gen-posix-lock-obj helper program: $ riscv64-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I../../src -I.. -Wdate-time -D_FORTIFY_SOURCE=2 -g -O2 -fdebug-prefix-map=/usr/src/repositories/libgpg-error=. -specs=/usr/share/dpkg/pie-compile.specs -fstack-protector-strong -Wformat -Werror=format-security -Wall -Wpointer-arith -Wno-psabi -fvisibility=hidden -c -o gen-posix-lock-obj.o ../../src/gen-posix-lock-obj.c $ /bin/bash ../libtool --tag=CC --mode=link riscv64-linux-gnu-gcc -g -O2 -fdebug-prefix-map=/usr/src/repositories/libgpg-error=. -specs=/usr/share/dpkg/pie-compile.specs -fstack-protector-strong -Wformat -Werror=format-security -Wall -Wpointer-arith -Wno-psabi -fvisibility=hidden -specs=/usr/share/dpkg/pie-link.specs -Wl,-z,relro -Wl,-z,now -o gen-posix-lock-obj gen-posix-lock-obj.o libtool: link: riscv64-linux-gnu-gcc -g -O2 -fdebug-prefix-map=/usr/src/repositories/libgpg-error=. -specs=/usr/share/dpkg/pie-compile.specs -fstack-protector-strong -Wformat -Werror=format-security -Wall -Wpointer-arith -Wno-psabi -fvisibility=hidden -specs=/usr/share/dpkg/pie-link.specs -Wl,-z -Wl,relro -Wl,-z -Wl,now -o gen-posix-lock-obj gen-posix-lock-obj.o $ ./gen-posix-lock-obj # (executed in qemu via binfmt-misc) ## lock-obj-pub.riscv64-unknown-linux-gnu.h ## File created by gen-posix-lock-obj - DO NOT EDIT ## To be included by mkheader into gpg-error.h typedef struct { long _vers; union { volatile char _priv[40]; long _x_align; long *_xp_align; } u; } gpgrt_lock_t; #define GPGRT_LOCK_INITIALIZER {1,{{0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0}}} ## ## Local Variables: ## mode: c ## buffer-read-only: t ## End: ## Attached is the raw header file as well as a patch for the packaging to carry this header file as a Debian patch. I would apprechiate very much if you could upload a new package version that includes this patch to unstable in the near future as we depend on it for continuing our bootstrap efforts. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. ## lock-obj-pub.riscv64-unknown-linux-gnu.h ## File created by gen-posix-lock-obj - DO NOT EDIT ## To be included by mkheader into gpg-error.h typedef struct { long _vers; union { volatile char _priv[40]; long _x_align; long *_xp_align; } u; } gpgrt_lock_t; #define GPGRT_LOCK_INITIALIZER {1,{{0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0, \ 0,0,0,0,0,0,0,0}}} ## ## Local Variables: ## mode: c ## buffer-read-only: t ## End: ## >From cec778b68abe56de49f838af5754348494778abe Mon Sep 17 00:00:00 2001 From: Karsten Merker <mer...@debian.org> Date: Tue, 27 Feb 2018 19:43:09 + Subject: [PATCH] add riscv64-unknown-linux-gnu lock-obj definitions --- ...bj-pub-file-for-riscv64-unknown-linux-gnu.patch | 30 ++ debian/patches/series | 1 + 2 files changed, 31 insertions(+) create mode 100644 debian/patches/0017-syscfg-Add-lock-obj-pub-file-for-riscv64-unknown-linux-gnu.patch diff --git a/debian/patches/0017-syscfg-Add-lock-obj-pub-file-for-riscv64-unknown-linux-gnu.patch b/debian/patches/0017-syscfg-Add-lock-obj-pub-file-for-riscv64-unknown-linux-gnu.patch new file mode 100644 index 000..19ec11a --- /dev/null +++ b/debian/patches/0017-syscfg-Add-lock-obj-pub-file-for-riscv64-unknown-linux-gnu.patch @@ -0,0 +1,30 @@ +Index: libgpg-error/src/syscfg/lock-obj-pub.riscv64-unknown-linux-gnu.h +=== +--- /dev/null libgpg-error/src/syscfg/lock-obj-pub.riscv64-unknown-linux-gnu.h +@@ -0,0 +1,25 @@ ++## lock-obj-pub.riscv64-unknown-linux-gnu.h ++## File created by gen-posix-lock-obj - DO NOT EDIT ++## To be included b
Bug#887139: d-i daily 2018-01-14 amd64 damages UEFI setup on Fujitsu Lifebook AH532
Package: installation-reports Version: 2.66 Severity: critical -- Package-specific info: Boot method: Netinstall CD image from USB stick Image version: https://cdimage.debian.org/cdimage/daily-builds/daily/arch-latest/amd64/jigdo-cd/debian-testing-amd64-netinst.jigdo dated 2018-01-14 04:22 Date: 2018-01-14 Machine: Fujitsu Lifebook AH532, first version (with BIOS version 1.09) Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[O] Configure network: [O] Detect CD: [O] Load installer modules: [O] Clock/timezone setup: [O] User/password setup:[O] Detect hard drives: [O] Partition hard drives: [O] Install base system:[O] Install tasks: [O] Install boot loader:[O] Overall install:[O] Problems: Short summary: The installation process completed successfully, but the UEFI firmware is broken afterwards; it is impossible to get into the UEFI setup, boot from any external device anymore or get the system to use the CSM to boot from a classical MBR. Detailed description: This was an installation in EFI mode (for the first time on this device). The system was previously running an older BIOS-mode installation without problems, but got a new, bigger harddisk and therefore required a new installation which was done in EFI mode (which is nowadays the default mode the installer starts in). The installation itself worked without problems and the resulting Debian system works fine, but it isn't possible anymore to get into the BIOS/UEFI setup of the machine nor boot from any other device. When powering on the system, the message ": Boot Menu, : BIOS Setup" appears (as always), pressing F2 results in a short "Please wait..." (als always), but then one doesn't get into the UEFI/BIOS setup but instead the system directly boots GRUB. Pressing F12 for the boot menu results in a screen with a menu with the following structure: - Boot Menu debian - Application Menu Diagnostic Screen "Diagnostic Screen" just shows the Firmware splash screen: "Phoenix SecureCore Tiano(TM) Copyright 1985-2011 Phoenix Technologies Ltd. All Rights Reserved Fixed Disk: ATAPI CDROM: Press any key to exit" and the "debian" boot menu entry obviously starts GRUB. There is no way to make the system boot from any external device anymore, so it is impossible to boot a rescue system or reinstall the device from scratch. Fallback to BIOS-mode booting also doesn't work any more. Plugging back the the old harddisk with a BIOS-mode install results in a non-booting system. The UEFI/BIOS always tries to boot the "debian" EFI entry which for obvious reasons doesn't work with a BIOS-mode setup on the disk. There is also a Window 7 UEFI-mode installation on the (new) disk, which can be started from GRUB, but it doesn't show up any more in the firmware-provided boot-menu (which it did before installing Debian). Running "efibootmgr -v" from the installed systems results in "No BootOrder is set; firmware will attempt recovery", which is what doesn't seem to work. It somewhat looks like the UEFI/BIOS setup, the Firmware-level boot menu and the CSM for BIOS-mode booting are EFI applications that cannot be started anymore because some NVRAM entry is broken. BIOS Information from dmidecode: Vendor: FUJITSU // Phoenix Technologies Ltd. Version: Version 1.09 Release Date: 05/22/2012 Address: 0xE Runtime Size: 128 kB ROM Size: 4096 kB Characteristics: PCI is supported PC Card (PCMCIA) is supported PNP is supported BIOS is upgradeable BIOS shadowing is allowed Boot from CD is supported Selectable boot is supported EDD is supported 3.5"/720 kB floppy services are supported (int 13h) Print screen service is supported (int 5h) 8042 keyboard services are supported (int 9h) Serial services are supported (int 14h) Printer services are supported (int 17h) CGA/mono video services are supported (int 10h) ACPI is supported USB legacy is supported BIOS boot specification is supported Function key-initiated network boot is supported Targeted content distribution is supported UEFI is supported BIOS Revision: 1.9 This is the newest BIOS version that is available for this device; newer versions (2.xx) are only available for a newer board revision and are not compatible with this board revision. How can one bring the NVRAM back into a sane state that allows getting into the setup and booting from external devices? Regards, Karsten Technical information about the system:
Bug#885712: libdebian-installer should not use -Werror
Control: tag 885712 pending On Fri, Dec 29, 2017 at 01:36:47PM +0100, Helmut Grohne wrote: > Source: libdebian-installer > Version: 0.112 > Severity: wishlist > Tags: patch > User: helm...@debian.org > Usertags: rebootstrap > > The packaging of libdebian-installer insterts a -Werror into CFLAGS. > This has caused FTBFS with gcc-6 and gcc-7 already and will cause more > FTBFS with gcc-8. Given that these issues are not fixed proactively, it > complicates architecture bootstrap, which somestimes has to use > unreleased compilers. I thus ask you to disable use of -Werror. This > avoids unexpected FTBFS. I still recommend doing maintainer builds with > -Werror to catch issues early. Just save everyone else from the fallout > please. > > Helmut > diff --minimal -Nru libdebian-installer-0.112/debian/changelog > libdebian-installer-0.112+nmu1/debian/changelog > --- libdebian-installer-0.112/debian/changelog2017-11-19 > 18:12:25.0 +0100 > +++ libdebian-installer-0.112+nmu1/debian/changelog 2017-12-29 > 13:32:02.0 +0100 > @@ -1,3 +1,10 @@ > +libdebian-installer (0.112+nmu1) UNRELEASED; urgency=medium > + > + * Non-maintainer upload. > + * Do not compile with -Werror by default. Closes: #-1. > + > + -- Helmut GrohneFri, 29 Dec 2017 13:32:02 +0100 > + > libdebian-installer (0.112) unstable; urgency=medium > >[ Reiner Herrmann ] > diff --minimal -Nru libdebian-installer-0.112/debian/rules > libdebian-installer-0.112+nmu1/debian/rules > --- libdebian-installer-0.112/debian/rules2017-11-19 17:26:42.0 > +0100 > +++ libdebian-installer-0.112+nmu1/debian/rules 2017-12-29 > 13:31:59.0 +0100 > @@ -6,7 +6,7 @@ > DEB_HOST_MULTIARCH := $(shell dpkg-architecture -qDEB_HOST_MULTIARCH) > > #CFLAGS = -Wall -W -Werror -ggdb -Wstrict-prototypes -Wmissing-declarations > -Wmissing-prototypes > -CFLAGS = -Wall -W -Werror -ggdb -Wmissing-declarations > +CFLAGS = -Wall -W -ggdb -Wmissing-declarations > > ifneq (,$(findstring noopt,$(DEB_BUILD_OPTIONS))) > CFLAGS += -O0 Hello, the patch has been applied to the libdebian-installer git repository. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#864562: u-boot: broken ethernet support on Olimex A20-Olinuxino-Micro Rev. J
control: reassign 864562 u-boot-sunxi control: retitle 864562 u-boot: broken ethernet support on Olimex A20-Olinuxino-Micro Rev. J control: tags 864562 patch On Mon, Oct 16, 2017 at 10:45:57PM +0200, Karsten Merker wrote: > On Mon, Oct 16, 2017 at 09:39:36PM +0200, Jean-Louis Mounier wrote: [Ethernet not working on Olimex A20-Olinuxino Micro Rev. J] > > I discovered a new installer release but the bug is still > > present. > > > > Now the test : > > > > => gpio clear PA17 > > gpio: pin PA17 (gpio 17) value is 0 > > =>run bootcmd > > > > after that, the installer got IPV4 and IPV6 addresses wwith DHCP and then > > ran fine. > > > > Hope it helps > > Yes, that helps as it confirms that a change in u-boot alone is > enough to work around the issue in Linux as well (although it is > debateable whether the kernel should rely on the bootloader in > this case or not). There has recently been an attempt to address > the issue in mainline u-boot, but that didn't work out as planned > and the patch got rejected: > > https://lists.denx.de/pipermail/u-boot/2017-September/307778.html Hello, a patch to address this issue has finally been accepted today by the upstream u-boot custodian for the sunxi platform, targeting u-boot v2018.01: https://patchwork.ozlabs.org/patch/833763/ Vagrant, could you include the patch in the next u-boot package upload? Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#881969: making bootable SD cards
Control: severity 881969 wishlist [CCing Vagrant Cascadian, the Debian u-boot maintainer] On Thu, Nov 16, 2017 at 07:54:42PM -0400, Joey Hess wrote: > Package: flash-kernel > Version: 3.87 > Severity: normal > > Therefore you usually have to setup an SD card with the appropriate u-boot > version for your particular device (see below) as a prerequisite for > installing Debian. If you use the pre-made SD card images with the > installer, this step is not necessary, as these images already contain > u-boot. > -- https://wiki.debian.org/InstallingDebianOn/Allwinner > > This seems to fall squarely in flash-kernel's wheelhouse, since it > already handles the other parts of u-boot installation, and it knows > the name of the board being installed. > > The d-i SD card images avoid the problem, but they are only one way to > install; there are even other ways to use d-i to install that need this > manual step first. > > The main difficulty in putting it in flash-kernel is that it might be > installed in a chroot in a situation where it should not be altering > the boot sector of the host's disk. So, something like grub-installer > seems to be called for, so the user specifies the device to install to. > > A utility in flash-kernel would be much nicer than needing to puzzle out dd > commands from README.Debian files and hope you got it right. I'm currently > having to embed those dd commands inside propellor; they're also embedded > inside debian-installer (build/boot/arm/u-boot-image-config). Hello, to use d-i/flash-kernel on the target device, one obviously needs to already have put a u-boot onto the device in some form (such as preinstalled in the d-i SD card images), otherwise one couldn't have booted it, i.e. flash-kernel could only cover the topic of updating u-boot from within a running system. There has been a discussion about the latter in the past and the consensus reached at that time was that it is practically impossible to safely determine the version of an already installed u-boot image in a platform-independant way, and installing u-boot unconditionally on every invocation of flash-kernel is considered too riscy as there are a number of unsolved problems with such an approach. Among the points of this discussion were: - On some devices u-boot isn't stored on an SD card but on an onboard SPI flash chip with a rather limited number of write cycles (in the area of ~1000) and no defects management. Unconditionally writing u-boot on every invocation of flash-kernel (which includes everything that modifies the initrd) would have the potential to kill these devices in comparatively short time. - Knowing the device type one is running on isn't necessarily enough. Several supported devices are available in different hardware configuration variants that influence where the u-boot image can get written to (SD card, onboard eMMC, onboard raw NAND, SPI flash, and combinations thereof). Once Linux is running, there is no way to determine where the u-boot that brought the system up was loaded from. Flash-kernel pulls the system type from a /proc entry, but that doesn't provide the information whether the current device e.g. has the SPI flash for u-boot populated or not, and if yes, whether it has actually been used for booting the system, so flash-kernel cannot decide without user-interaction where to write the u-boot image. - As u-boot is more than just a bootloader - it also provides BIOS-like functionality - there can be a major difference between messing up automatically installing GRUB and messing up automatically installing u-boot. In the GRUB case, the user can simply boot a rescue system and fix the bootloader. In case of a broken u-boot installation to an SPI flash or to an eMMC on systems where these have a higher boot priority than the SD slot, the system can be completely dead and require specific hardware tooling (such as an external SPI flasher) to revive the system again. As a result of these issues, it was deemed unsuitable for flash-kernel to attempt installing u-boot. What we might do sometime in the future is adding a u-boot-installer udeb to d-i which on a very limited subset of systems allows the user to explicitly decide to install u-boot to a user-selected device (such as eMMC or SPI flash) after being informed about the riscs of doing so. I had started designing a proof-of-concept for such a udeb, but have had to put that on hold due to having to take care of a number of higher-priority issues. I'm setting the severity of this bug down from "normal" to "wishlist" as it is about requesting the addition of a new feature and not about a bug in existing functionality. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#868166: flash-kernel: Patch for Radxa Rock2 Square support
Control: tags 868166 pending On Wed, Jul 12, 2017 at 03:57:13PM +, Eva Ramon wrote: > Package: flash-kernel > Version: 3.82 > Severity: wishlist > > Dear Maintainer, > > please apply the patch provided in order to add support for the Radxa Rock2 > Square board to flash-kernel. > > --- /usr/share/flash-kernel/db/all.db.old 2017-07-12 15:08:15.380566406 > + > +++ /usr/share/flash-kernel/db/all.db 2017-07-12 15:08:34.300272594 + > @@ -378,6 +378,13 @@ > U-Boot-Script-Name: bootscr.uboot-generic > Required-Packages: u-boot-tools > > +Machine: Radxa Rock 2 Square > +Kernel-Flavors: armmp armmp-lpae > +DTB-Id: rk3288-rock2-square.dtb > +Boot-Script-Path: /boot/boot.scr > +U-Boot-Script-Name: bootscr.uboot-generic > +Required-Packages: u-boot-tools > + > Machine: Freescale i.MX53 Quick Start Board > Machine: Freescale MX53 LOCO Board > Kernel-Flavors: armmp mx5 > > Thanks! Hello, many thanks for the patch submission. The machine db entry has been added to the flash-kernel git repository and will enter the archive on the next package upload (probably in a few days). Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#866521: [PATCH] initramfs-tools: rootfs on USB mass storage broken for a number of armhf systems
Package: initramfs-tools Severity: important Tags: patch Hello, with newer kernels (including kernel 4.9 as used in stretch), initializing the USB host controllers on a number of armhf systems is not possible without having the "axp20x_usb_power" power supply driver module available. To be able to have the rootfs on a USB mass storage device on such systems it is therefore necessary to include this module in the initramfs, which currently isn't the case. Attached is a patch that adds the "axp20x_usb_power" module to the "base" module list for the "MODULES=most" setting. JFTR: The patch doesn't address the "MODULES=dep" case; rootfs on USB mass storage currently doesn't work at all with "MODULES=dep" as the necessary USB controller drivers don't get included. An Example (Allwinner A20) with MODULES=dep: lib/modules/4.9.0-3-armmp-lpae/kernel/drivers lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/usb lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/usb/common lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/usb/common/usb-common.ko lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/usb/phy lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/usb/phy/phy-generic.ko lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/scsi lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/scsi/sd_mod.ko lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/scsi/scsi_mod.ko lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/i2c lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/i2c/busses lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/i2c/busses/i2c-mv64xxx.ko lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/phy lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/phy/phy-sun4i-usb.ko lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/extcon lib/modules/4.9.0-3-armmp-lpae/kernel/drivers/extcon/extcon-core.ko Missing are at least ehci_platform, ehci_hcd, ohci_platform, ohci_hcd and usbcore (when using the OTG controller also musb_hdrc and udc_core) plus the axp20x_regulator regulator driver and the axp20x_usb_power power supply driver. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. >From 7358533bcf35c7237ad8ab57dd2fe0b7db5f553c Mon Sep 17 00:00:00 2001 From: Karsten Merker <mer...@debian.org> Date: Thu, 29 Jun 2017 20:51:03 +0200 Subject: [PATCH] Include the axp20x_usb_power module in the base modules list for MODULES=most With newer kernels (including kernel 4.9 as used in stretch), initializing the USB host controllers on a number of armhf systems is not possible without having the "axp20x_usb_power" power supply driver module available. To be able to have the rootfs on a USB mass storage device on such systems it is therefore necessary to include this module in the initramfs. --- hook-functions | 4 1 file changed, 4 insertions(+) diff --git a/hook-functions b/hook-functions index 679e11d..4c68c90 100644 --- a/hook-functions +++ b/hook-functions @@ -513,6 +513,10 @@ auto_add_modules() copy_modules_dir kernel/drivers/usb/renesas_usbhs # and any extcon drivers for USB modules="$modules extcon-usb-gpio" + # Add the axp20x_usb_power power supply driver, + # required to initialize the USB host controllers + # on a number of armhf systems + modules="$modules axp20x_usb_power" # Include all HID drivers unless we're sure they # don't support keyboards. hid-*ff covers various -- 2.13.2
Bug#864457: [d-i daily hd-media 20170608] kernel doesn't initialize USB ports inside d-i on a sunxi-based system
On Thu, Jun 08, 2017 at 10:50:57PM +0200, Karsten Merker wrote: > Package: installation-reports > > Boot method: hd-media tarball from USB stick > Image version: > https://d-i.debian.org/daily-images/armhf/daily/hd-media/hd-media.tar.gz > Date: 2017-06-08 > > Machine: LeMaker Banana Pi > Processor: Allwinner A20 (2x Cortex A7) > Memory: 1GB [...] > Comments/Problems: > Installation method is a d-i hd-media tarball and a Debian CD/DVD > ISO image on a USB stick. Booting d-i from the stick works fine, > but inside the installer the USB ports are dead and no USB > devices get recognized by the kernel. > > In an installed system on the same hardware (installed with the > netboot image), the USB ports work normally. In the d-i > environment, the ehci platform driver gets loaded, but for some > reason doesn't initialize the actual host controller. All > relevant USB host driver modules appear to be there: > > Module Size Used by > usb_storage45835 0 > phy_generic 4686 1 sunxi > musb_hdrc 113325 1 sunxi > udc_core 26444 1 musb_hdrc > ohci_platform 4786 0 > ohci_hcd 37898 1 ohci_platform > ehci_platform 5462 0 > ehci_hcd 64996 1 ehci_platform > phy_sun4i_usb 8637 1 sunxi > extcon_core13223 2 sunxi,phy_sun4i_usb > usbcore 196310 6 > usb_storage,ehci_hcd,musb_hdrc,ohci_hcd,ehci_platform,ohci_platform > usb_common 3659 5 udc_core,sunxi,musb_hdrc,phy_sun4i_usb,usbcore > > In the d-i environment: > [2.694030] usbcore: registered new interface driver usbfs > [2.699958] usbcore: registered new interface driver hub > [2.703305] SCSI subsystem initialized > [2.716192] usbcore: registered new device driver usb > [2.724836] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > [2.736947] sunxi-mmc 1c0f000.mmc: base:0xf08f4000 irq:28 > [2.750665] ehci-platform: EHCI generic platform driver > [2.751130] ohci_hcd: USB 1.1 'Open' Host Controller (OHCI) Driver > [2.752551] ohci-platform: OHCI generic platform driver > > While on the installed system: > [2.162185] ehci_hcd: USB 2.0 'Enhanced' Host Controller (EHCI) Driver > [2.168314] ehci-platform: EHCI generic platform driver > [6.187194] ehci-platform 1c14000.usb: EHCI Host Controller > [6.187256] ehci-platform 1c14000.usb: new USB bus registered, assigned > bus number 2 > [6.192414] ehci-platform 1c14000.usb: irq 30, io mem 0x01c14000 > [6.205562] ehci-platform 1c14000.usb: USB 2.0 started, EHCI 1.00 > [6.206170] usb usb2: New USB device found, idVendor=1d6b, idProduct=0002 > [6.206184] usb usb2: New USB device strings: Mfr=3, Product=2, > SerialNumber=1 > [6.206191] usb usb2: Product: EHCI Host Controller > [6.206198] usb usb2: Manufacturer: Linux 4.9.0-3-armmp-lpae ehci_hcd > [6.206204] usb usb2: SerialNumber: 1c14000.usb > [6.207443] hub 2-0:1.0: USB hub found > [6.207557] hub 2-0:1.0: 1 port detected > [6.209230] ehci-platform 1c1c000.usb: EHCI Host Controller > [6.209289] ehci-platform 1c1c000.usb: new USB bus registered, assigned > bus number 3 > [6.216403] ehci-platform 1c1c000.usb: irq 34, io mem 0x01c1c000 > [6.230524] ehci-platform 1c1c000.usb: USB 2.0 started, EHCI 1.00 > [6.231044] usb usb3: New USB device found, idVendor=1d6b, idProduct=0002 > [6.231058] usb usb3: New USB device strings: Mfr=3, Product=2, > SerialNumber=1 > [6.231066] usb usb3: Product: EHCI Host Controller > [6.231072] usb usb3: Manufacturer: Linux 4.9.0-3-armmp-lpae ehci_hcd > [6.231079] usb usb3: SerialNumber: 1c1c000.usb > > We might have a problem with the USB port power supply here. > Comparing the available modules between the installed system and > the d-i environment (full list below) and looking for possibly > related modules shows that the "axp20x_regulator" and > "axp20x_usb_power" modules are available in the installed system, > but not in the d-i environment. I am not sure whether > axp20x_usb_power is really responsible for providing power _to_ > the USB hosts ports, though. It could be that it is responsible > for powering the system _from_ a USB-OTG port, so it might be > unrelated to the problem at hand. "axp20x_regulator" would > possibly be a candidate for a power-supply problem. I have also > tried using a powered USB hub between the USB stick and the host > port, but even then no USB device gets recognized, which kind of > speaks against the theory of only a missing 5V-supply on the USB > port as the sole source of the problem. Possibly
Bug#851469: flash-kernel: ARM new boot.scr does not allow the device to boot
On Sun, Jan 15, 2017 at 12:02:56PM +0100, permondes - sagen wrote: > Am Sonntag, den 15.01.2017, 11:47 +0100 schrieb Karsten Merker: > > This points at a problem with either your u-boot version or your > > u-boot environment. The oldest u-boot version that Debian has > > ever shipped for the Olinuxino A20 LIME as part of the official > > Debian-Installer images has been mainline u-boot v2014.10, and > > all mainline u-boot versions since v2014.10 provide a default > > environment that is compatible with this boot script. > > > > Please note that u-boot keeps its old environment on upgrade by > > default, i.e. if you originally had e.g. some vendor-specific > > u-boot installed and upgraded to mainline u-boot later, this > > would have kept the environment from the vendor u-boot. You can > > reset the u-boot environment to its current default by running > > "env default -a; saveenv" at the u-boot command prompt. > > > > HTH, > > Karsten > > The device is being used as a FreedomBox, a flavour of Debian. They > provide an initial "live" image which is quite old actually. Updating to > the current version, which is done automatically during set-up, makes > this change of boot.scr, so I guess they switched to standard u-boot. Not necessarily. The boot.scr is generated by the flash-kernel package on every kernel update, but the u-boot in the boot area is not automatically updated with an update of the u-boot-sunxi package, so unless the people who have built the freedombox upgrade mechanism have explicitly implemented a function to update the u-boot in the boot area (which I doubt), you will probably still have whatever older u-boot version was used when building the original freedombox image. > I would like to test the reset of the environment, how do I get into > u-boot prompt? I just have ssh access to the box, no serial, nor > keyboard or so. Hm, that poses a problem. The "normal" way is via serial console, or, if the u-boot version is relatively new, via a display connected to the HDMI port and a USB keyboard. U-Boot isn't memory resident (except for the PSCI functions), i.e. once the Linux kernel is booted, there is no u-boot anymore with which one could interact directly. What you could try is clobbering the area on the SD card in which u-boot stores the environment. This results in a checksum mismatch and u-boot falls back to the integrated default environment. On the LIME, the u-boot environment is stored on the SD card starting at offset 544kB from the start of the device, and is 128kB in size. So on the LIME running $ sudo dd if=/dev/zero bs=1k count=128 seek=544 of=/dev/mmcblk0 should overwrite the complete environment area with zeros, resulting in u-boot falling back to the built-in defaults. But without some form of console you still won't be able to see which version of u-boot is really installed (cf. my mail to debian-arm) and what exactly happens at boot time. If you want to make sure that you have a current u-boot version installed, you can replace whatever version is currently on the SD card by running $ sudo dd if=/usr/lib/u-boot/A20-OLinuXino-Lime/u-boot-sunxi-with-spl.bin of=/dev/mmcblk0 bs=1k seek=8 Doing all this "blind" without console is of course inherently dangerous, so make an image backup of your SD card that you can restore in case of problems before trying any of this. HTH, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#851469: flash-kernel: ARM new boot.scr does not allow the device to boot
On Sun, Jan 15, 2017 at 10:45:29AM +0100, permondes - sagen wrote: > Package: flash-kernel > Version: 3.73 > Severity: important > > Dear Maintainer, > > I am using an OLinuXino A20 LIME with arm. > boot.scr used to have the following content (I removed the initial > non-readable characters) which let boot the device without issues: > > setenv mmcpart 1 > > setenv mmcroot /dev/mmcblk0p2 ro > setenv mmcrootfstype btrfs rootwait fixrtc > setenv mmcrootflags subvol=@ > > setenv console ttyS0,115200n8 > > setenv kernel_file vmlinuz-4.8.0-2-armmp-lpae > setenv initrd_file initrd.img-4.8.0-2-armmp-lpae > setenv fdtfile sun7i-a20-olinuxino-lime.dtb > > setenv loadaddr 0x4600 > setenv initrd_addr 0x4800 > setenv fdtaddr 0x4700 > > setenv initrd_high 0x > setenv fdt_high 0x > > setenv loadkernel load mmc ${mmcdev}:${mmcpart} ${loadaddr} > ${kernel_file} > setenv loadinitrd load mmc ${mmcdev}:${mmcpart} ${initrd_addr} > ${initrd_file}\; setenv initrd_size \${filesize} > setenv loadfdt load mmc ${mmcdev}:${mmcpart} ${fdtaddr} /dtbs/${fdtfile} > > setenv loadfiles run loadkernel\; run loadinitrd\; run loadfdt > setenv mmcargs setenv bootargs console=${console} root=${mmcroot} > rootfstype=${mmcrootfstype} rootflags=${mmcrootflags} > > run loadfiles; run mmcargs; bootz ${loadaddr} ${initrd_addr}: > ${initrd_size} ${fdtaddr} > > <-- end of file To my knowledge Debian has never provided a boot script with this content as a part of its official flash-kernel releases. Is this perhaps from some non-official Debian image that somebody else has produced, e.g. from some image provided by the hardware manufacturer or from some other project that uses Debian as its base but doesn't use Debian's normal mechanisms for handling the boot scripts? Which u-boot version do you have installed? Can you provide the output of the "printenv" command at the u-boot prompt? > The new content of that file is: > > # boot script for Allwinner SunXi-based devices > > # Mainline u-boot v2014.10 introduces a new default environment and > # a new common bootcmd handling for all platforms, which is not fully > # compatible with the old-style environment used by u-boot-sunxi. > # This script therefore needs to check in which environment it > # is running and set some variables accordingly. > > # On u-boot-sunxi, this script assumes that ${device} and ${partition} > # are set. > > # The new-style environment predefines ${boot_targets}, the old-style > # environment does not. > if test -n "${boot_targets}" > then > echo "Mainline u-boot / new-style environment detected." > # Mainline u-boot v2014.10 uses ${devtype}, ${devnum} and > # ${bootpart} where u-boot-sunxi uses ${device} and ${partition}. > # ${distro_bootpart} replaced ${bootpart} in u-boot v2016.01. > if test -z "${device}"; then setenv device "${devtype}"; fi > if test -z "${partition}${distro_bootpart}"; then setenv partition > "${devnum}:${bootpart}"; fi > if test -z "${partition}"; then setenv partition "${devnum}: > ${distro_bootpart}"; fi > else > echo "U-boot-sunxi / old-style environment detected." > # U-boot-sunxi does not predefine kernel_addr_r, fdt_addr_r and > # ramdisk_addr_r, so they have to be manually set. Use the values > # from mainline u-boot v2014.10, except for ramdisk_addr_r, > # which is set to 0x4430 to allow for initrds larger than > # 13MB on u-boot-sunxi. > setenv kernel_addr_r 0x4200 > setenv fdt_addr_r 0x4300 > setenv ramdisk_addr_r 0x4430 > fi > > if test -n "${console}"; then > setenv bootargs "${bootargs} console=${console}" > fi > > setenv bootargs ${bootargs} quiet > > > image_locations='/boot/ /' > if test -z "${fk_kvers}"; then >setenv fk_kvers '4.8.0-2-armmp-lpae' > fi > > if test -n "${fdtfile}"; then >setenv fdtpath dtbs/${fk_kvers}/${fdtfile} > else >setenv fdtpath dtb-${fk_kvers} > fi > > for pathprefix in ${image_locations} > do > if test -e ${device} ${partition} ${pathprefix}vmlinuz-${fk_kvers} > then > load ${device} ${partition} ${kernel_addr_r} > ${pathprefix}vmlinuz-${fk_kvers} \ > && load ${device} ${partition} ${fdt_addr_r} ${pathprefix}${fdtpath} > \ > && load ${device} ${partition} ${ramdisk_addr_r} > ${pathprefix}initrd.img-${fk_kvers} \ > && echo "Booting Debian ${fk_kvers} from ${device} ${partition}..." > \ > && bootz ${kernel_addr_r} ${ramdisk_addr_r}:${filesize} > ${fdt_addr_r} > fi > done > > <- end of file > > Which does not allow the device to boot. > None of the variables in that script is set in the current environment. This points at a problem with either your u-boot version or your u-boot environment. The oldest u-boot version that Debian has ever shipped for the Olinuxino A20 LIME as part of the official Debian-Installer images has been mainline u-boot v2014.10, and all mainline u-boot versions since v2014.10 provide a default environment that is compatible with this boot script.
Bug#830169: u-boot 2016.07~rc3+dfsg1-1: FTBFS for openrd_base, openrd_client and openrd_ultimate targets
Source: u-boot Version: 2016.07~rc3+dfsg1-1 Severity: important Tags: upstream U-Boot v2016.07~rc3 fails to build from source for the (armel) targets openrd_base, openrd_client and openrd_ultimate The problem is not specific to the Debian package but exists in upstream git as well. A bug report has been sent to the upstream u-boot development list. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung.
Bug#777334: [jessie daily 2015-02-07] [armhf] installation-report: A20-OLinuXino-LIME
Package: installation-reports Boot method: hd-media + CD image on a USB stick hd-media version: http://d-i.debian.org/daily-images/armhf/20150207-05:30/hd-media/hd-media.tar.gz CD version: Debian GNU/Linux testing Jessie - Official Snapshot armhf CD Binary-1 20150202-10:45 from http://cdimage.debian.org/cdimage/weekly-builds/armhf/jigdo-cd/debian-testing-armhf-CD-1.jigdo, dated 2015-02-02 12:06 Date: 2015-02-07 Machine: Olimex A20-OLinuXino-LIME Processor: Allwinner A20 (2x Cortex A7) Memory: 512MB Console: serial console, 115200 baud Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[O] Configure network: [O] Detect CD: [O] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [O] Install base system:[O] Clock/timezone setup: [O] User/password setup:[O] Install tasks: [O] Install boot loader:[O] Overall install:[O] Comments/Problems: The installation was done to a 16GB Micro-SD card using serial console (on the UART0 header) and worked without problems; network (100MBit) and the USB host ports work as expected. Partitioning was done using the guided partitioning, all files in one partition option. I have not tested using a SATA disk with the board due to lack of a SATA power cable that fits into the board's SATA power connector (2-pin JST-style). As an informational note: due to bug #773645 (lack of u-boot tools on the first armhf CD), hd-media/CD installations on armhf platforms currently require network connectivity during the installation process; pure offline installations are not possible with hd-media/CD at the moment. dmesg output: [0.00] Booting Linux on physical CPU 0x0 [0.00] Initializing cgroup subsys cpuset [0.00] Initializing cgroup subsys cpu [0.00] Initializing cgroup subsys cpuacct [0.00] Linux version 3.16.0-4-armmp-lpae (debian-ker...@lists.debian.org) (gcc version 4.8.3 (Debian 4.8.3-16) ) #1 SMP Debian 3.16.7-ckt2-1 (2014-12-08) [0.00] CPU: ARMv7 Processor [410fc074] revision 4 (ARMv7), cr=30c5387d [0.00] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache [0.00] Machine model: Olimex A20-OLinuXino-LIME [0.00] Forcing write-allocate cache policy for SMP [0.00] Memory policy: Data cache writealloc [0.00] On node 0 totalpages: 131072 [0.00] free_area_init_node: node 0, pgdat c0a0cb00, node_mem_map dfbf9000 [0.00] DMA zone: 1024 pages used for memmap [0.00] DMA zone: 0 pages reserved [0.00] DMA zone: 131072 pages, LIFO batch:31 [0.00] psci: probing for conduit method from DT. [0.00] psci: Using PSCI v0.1 Function IDs from DT [0.00] PERCPU: Embedded 9 pages/cpu @dfbd4000 s13312 r8192 d15360 u36864 [0.00] pcpu-alloc: s13312 r8192 d15360 u36864 alloc=9*4096 [0.00] pcpu-alloc: [0] 0 [0] 1 [0.00] Built 1 zonelists in Zone order, mobility grouping on. Total pages: 130048 [0.00] Kernel command line: console=ttyS0,115200 quiet [0.00] PID hash table entries: 2048 (order: 1, 8192 bytes) [0.00] Dentry cache hash table entries: 65536 (order: 6, 262144 bytes) [0.00] Inode-cache hash table entries: 32768 (order: 5, 131072 bytes) [0.00] Memory: 497040K/524288K available (6562K kernel code, 838K rwdata, 2216K rodata, 701K init, 396K bss, 27248K reserved, 0K highmem) [0.00] Virtual kernel memory layout: vector : 0x - 0x1000 ( 4 kB) fixmap : 0xffc0 - 0xffe0 (2048 kB) vmalloc : 0xe080 - 0xff00 ( 488 MB) lowmem : 0xc000 - 0xe000 ( 512 MB) pkmap : 0xbfe0 - 0xc000 ( 2 MB) modules : 0xbf00 - 0xbfe0 ( 14 MB) .text : 0xc0008000 - 0xc089aeb8 (8780 kB) .init : 0xc089b000 - 0xc094a400 ( 701 kB) .data : 0xc094c000 - 0xc0a1dab8 ( 839 kB) .bss : 0xc0a1dab8 - 0xc0a80eb4 ( 397 kB) [0.00] Hierarchical RCU implementation. [0.00] RCU dyntick-idle grace-period acceleration is enabled. [0.00] RCU restricting CPUs from NR_CPUS=4 to nr_cpu_ids=2. [0.00] RCU: Adjusting geometry for rcu_fanout_leaf=16, nr_cpu_ids=2 [0.00] NR_IRQS:16 nr_irqs:16 16 [0.00] Architected cp15 timer(s) running at 24.00MHz (phys). [0.09] sched_clock: 56 bits at 24MHz, resolution 41ns, wraps every 2863311519744ns [0.19] Switching to timer-based delay loop [0.001002] sched_clock: 32 bits at 24MHz, resolution 41ns, wraps every 178956969942ns [0.001363] sched_clock: 32 bits at 160MHz, resolution 6ns, wraps every 26843545593ns [0.001801] Console: colour dummy device 80x30 [0.001837] Calibrating delay loop (skipped), value calculated using timer frequency.. 48.00 BogoMIPS (lpj=96000) [
Bug#774067: linux 3.16.7-ckt: [patch] please add a dts for the LinkSprite pcDuino V3
Source: linux Version: 3.16.7-ckt2-1 Severity: wishlist Tags: patch On the debian-arm mailinglist, a user has asked for support for the LinkSprite pcDuino V3 board (see the thread starting at https://lists.debian.org/debian-arm/2014/12/msg00052.html). Driver-wise all core components of the board (serial console, MMC, SATA, USB, wired ethernet) are supported by kernel 3.16, it just lacks an appropriate device-tree file, which was added upstream in kernel 3.17. Attached is a patch to backport the device-tree file. As this is a completely self-contained addition and does not bring any code changes, I believe this to be suitable for inclusion into Jessie despite the freeze. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. Index: debian/patches/features/arm/dts-sun7i-Add-board-support-for-LinkSprite-pcDuino-V3.patch === --- debian/patches/features/arm/dts-sun7i-Add-board-support-for-LinkSprite-pcDuino-V3.patch (revision 0) +++ debian/patches/features/arm/dts-sun7i-Add-board-support-for-LinkSprite-pcDuino-V3.patch (working copy) @@ -0,0 +1,206 @@ +From 04089927981f295b42cd695485383b2d11283d59 Mon Sep 17 00:00:00 2001 +From: Zoltan HERPAI wigy...@uid0.hu +Date: Mon, 30 Jun 2014 23:57:56 +0200 +Subject: ARM: dts: sun7i: Add board support for LinkSprite pcDuino V3 +Origin: https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/patch/?id=04089927981f295b42cd695485383b2d11283d59 + +The LinkSprite pcDuino V3 is an A20 based development board featuring +arduino compatible io headers, 1G RAM, 4G nand, sata, rtl8188cus usb wifi +and 100 Mbit ethernet using an ip101a phy: + +http://www.pcduino.com/pcduino-v3/ + +Signed-off-by: Zoltan HERPAI wigy...@uid0.hu +[hdego...@redhat.com: Various cleanups, correct led pins] +[hdego...@redhat.com: Add axp209, ir and gpio-keys nodes] +Signed-off-by: Hans de Goede hdego...@redhat.com +Signed-off-by: Maxime Ripard maxime.rip...@free-electrons.com + +--- a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile +@@ -381,7 +381,8 @@ + sun7i-a20-cubietruck.dtb \ + sun7i-a20-i12-tvbox.dtb \ + sun7i-a20-olinuxino-lime.dtb \ +- sun7i-a20-olinuxino-micro.dtb ++ sun7i-a20-olinuxino-micro.dtb \ ++ sun7i-a20-pcduino3.dtb + dtb-$(CONFIG_ARCH_TEGRA) += tegra20-harmony.dtb \ + tegra20-iris-512.dtb \ + tegra20-medcom-wide.dtb \ +--- /dev/null b/arch/arm/boot/dts/sun7i-a20-pcduino3.dts +@@ -0,0 +1,173 @@ ++/* ++ * Copyright 2014 Zoltan HERPAI ++ * Zoltan HERPAI wigy...@uid0.hu ++ * ++ * The code contained herein is licensed under the GNU General Public ++ * License. You may obtain a copy of the GNU General Public License ++ * Version 2 or later at the following locations: ++ * ++ * http://www.opensource.org/licenses/gpl-license.html ++ * http://www.gnu.org/copyleft/gpl.html ++ */ ++ ++/dts-v1/; ++/include/ sun7i-a20.dtsi ++/include/ sunxi-common-regulators.dtsi ++#include dt-bindings/gpio/gpio.h ++#include dt-bindings/input/input.h ++ ++/ { ++ model = LinkSprite pcDuino3; ++ compatible = linksprite,pcduino3, allwinner,sun7i-a20; ++ ++ soc@01c0 { ++ mmc0: mmc@01c0f000 { ++ pinctrl-names = default; ++ pinctrl-0 = mmc0_pins_a, mmc0_cd_pin_reference_design; ++ vmmc-supply = reg_vcc3v3; ++ bus-width = 4; ++ cd-gpios = pio 7 1 0; /* PH1 */ ++ cd-inverted; ++ status = okay; ++ }; ++ ++ usbphy: phy@01c13400 { ++ usb1_vbus-supply = reg_usb1_vbus; ++ usb2_vbus-supply = reg_usb2_vbus; ++ status = okay; ++ }; ++ ++ ehci0: usb@01c14000 { ++ status = okay; ++ }; ++ ++ ohci0: usb@01c14400 { ++ status = okay; ++ }; ++ ++ ahci: sata@01c18000 { ++ target-supply = reg_ahci_5v; ++ status = okay; ++ }; ++ ++ ehci1: usb@01c1c000 { ++ status = okay; ++ }; ++ ++ ohci1: usb@01c1c400 { ++ status = okay; ++ }; ++ ++ pinctrl@01c20800 { ++ ahci_pwr_pin_a: ahci_pwr_pin@0 { ++allwinner,pins = PH2; ++ }; ++ ++ led_pins_pcduino3: led_pins@0 { ++allwinner,pins = PH15, PH16; ++allwinner,function = gpio_out; ++allwinner,drive = 0; ++allwinner,pull = 0; ++ }; ++ ++ key_pins_pcduino3: key_pins@0 { ++allwinner,pins = PH17, PH18, PH19; ++allwinner,function = gpio_in; ++allwinner,drive = 0; ++allwinner,pull = 0; ++ }; ++ }; ++ ++ ir0: ir@01c21800 { ++ pinctrl-names = default; ++ pinctrl-0 = ir0_pins_a; ++ status = okay; ++ }; ++ ++ uart0: serial@01c28000 { ++ pinctrl-names = default; ++ pinctrl-0 = uart0_pins_a; ++ status = okay; ++ }; ++ ++ i2c0: i2c@01c2ac00 { ++ pinctrl-names = default; ++ pinctrl-0 = i2c0_pins_a; ++ status = okay; ++ ++ axp209: pmic@34 { ++compatible = x-powers,axp209; ++reg = 0x34; ++interrupt-parent = nmi_intc; ++interrupts = 0 8; ++ ++interrupt-controller; ++#interrupt-cells = 1; ++ }; ++ }; ++ ++ gmac:
Bug#773645: debian-cd: [armhf/armel/arm64] u-boot-tools missing on CD1 - d-i fails
Package: debian-cd Severity: important Hello, in the current daily CD builds, the first armhf CD does not include the package u-boot-tools. This package is required by d-i to make the system bootable at the end of the installation on nearly all armhf systems. As a consequence, single-CD installations, respectively hd-media installations that use the first CD iso, fail completely if no network access is available during the d-i run. From looking at the flash-kernel machine database, the situation will probably be the same for armel and arm64, so please include u-boot-tools on the first CD for armhf/armel/arm64. From the d-i log with CD1 but no network access: in-target: Setting up flash-kernel (3.28) ... in-target: in-target: Creating config file /etc/default/flash-kernel with new version^ in-target: Reading package lists... in-target: in-target: Building dependency tree... in-target: in-target: Reading state information... in-target: in-target: Package u-boot-tools is not available, but is referred to by another package. in-target: This may mean that the package is missing, has been obsoleted, or in-target: is only available from another source in-target: in-target: E: Package 'u-boot-tools' has no installation candidate flash-kernel-installer: error: apt-install u-boot-tools failed Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#773127: installation-reports: mostly Successful: Wandboard Quad [armhf]
On Sun, Dec 14, 2014 at 02:31:45PM -0800, Vagrant Cascadian wrote: On 2014-12-14, Karsten Merker wrote: Could you perhaps give a short list of which hardware components are working with Debian's kernel, so that I can add that to the installation guide as well? I understand from your installation report that MMC, LAN, USB and HDMI video work. Yes, those all work. [SATA and serial console ok, WLAN/Bluetooth and Audio not working] Thanks, I have added this information to the installation guide. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#773127: installation-reports: mostly Successful: Wandboard Quad [armhf]
On Sun, Dec 14, 2014 at 12:02:09PM -0800, Vagrant Cascadian wrote: Package: installation-reports Boot method: netboot Image version: http://d-i.debian.org/daily-images/armhf/20141214-05:27/netboot/ Date: 20141214 Machine: Wandboard Quad Installing with the console installer worked fine. Hello, based on your installation report, I have just added the Wandboard Quad to the list of supported systems in the installation guide. Could you perhaps give a short list of which hardware components are working with Debian's kernel, so that I can add that to the installation guide as well? I understand from your installation report that MMC, LAN, USB and HDMI video work. The module list shows modules loaded for SATA and WLAN, but I cannot assess from the module list whether these actually work correctly (on some systems there can be be issues with GPIO-controlled regulators for the SATA power or a requirement for OOB communication to the SDIO WLAN module), so I would appreciate an explicit confirmation. Besides these, do the following components work? - Serial Console - Audio (HDMI, analog, S/PDIF) - USB OTG - Bluetooth Kind Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#769728: [jessie daily 2014-11-15] installation-report: regression when deselecting the standard system utilities task
Package: installation-reports Boot method: netboot Image version: http://d-i.debian.org/daily-images/armhf/daily/netboot/ dated 15-Nov-2014 05:25 Date: 2014-11-15 Machine: Cubietech Cubietruck Processor: Allwinner A20 (2x Cortex A7) Memory: 2GB Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[O] Configure network: [O] Detect CD: [O] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [O] Install base system:[O] Clock/timezone setup: [O] User/password setup:[O] Install tasks: [O] Install boot loader:[O] Overall install:[E] Comments/Problems: The installation itself works flawlessly, but when installing a minimal system by deselecting standard system utilities in tasksel, on booting the installed system the systemd error message [FAILED] Failed to start Login Service gets shown several times. systemctl status -l systemd-logind.service results in: -8--8--8--8--8--8- systemd-logind.service - Login Service Loaded: loaded (/lib/systemd/system/systemd-logind.service; static) Active: failed (Result: start-limit) since Sat 2014-11-15 22:21:56 CET; 1min 13s ago Docs: man:systemd-logind.service(8) man:logind.conf(5) http://www.freedesktop.org/wiki/Software/systemd/logind http://www.freedesktop.org/wiki/Software/systemd/multiseat Process: 273 ExecStart=/lib/systemd/systemd-logind (code=exited, status=1/FAILURE) Main PID: 273 (code=exited, status=1/FAILURE) Status: Shutting down... Nov 15 22:21:56 debian systemd[1]: Failed to start Login Service. Nov 15 22:21:56 debian systemd[1]: Unit systemd-logind.service entered failed state. Nov 15 22:21:56 debian systemd[1]: systemd-logind.service has no holdoff time, scheduling restart. Nov 15 22:21:56 debian systemd[1]: Stopping Login Service... Nov 15 22:21:56 debian systemd[1]: Starting Login Service... Nov 15 22:21:56 debian systemd[1]: systemd-logind.service start request repeated too quickly, refusing to start. Nov 15 22:21:56 debian systemd[1]: Failed to start Login Service. Nov 15 22:21:56 debian systemd[1]: Unit systemd-logind.service entered failed state. -8--8--8--8--8--8- When standard system utilities is selected in tasksel, the problem does not occur. Deselecting standard system utilities did not have any negatve effect on the boot process when d-i still installed SysV-init as the default init system, so this is a regression. Following is a full boot log of an installation with standard system utilities deselected in tasksel: -8--8--8--8--8--8- Starting kernel ... [0.00] Booting Linux on physical CPU 0x0 [0.00] Initializing cgroup subsys cpuset [0.00] Initializing cgroup subsys cpu [0.00] Initializing cgroup subsys cpuacct [0.00] Linux version 3.16.0-4-armmp-lpae (debian-ker...@lists.debian.org) (gcc version 4.8.3 (Debian 4.8.3-13) ) #1 SMP Debian 3.16.7-2 (2014-11-06) [0.00] CPU: ARMv7 Processor [410fc074] revision 4 (ARMv7), cr=30c5387d [0.00] CPU: PIPT / VIPT nonaliasing data cache, VIPT aliasing instruction cache [0.00] Machine model: Cubietech Cubietruck [0.00] Forcing write-allocate cache policy for SMP [0.00] Memory policy: Data cache writealloc [0.00] psci: probing for conduit method from DT. [0.00] psci: Using PSCI v0.1 Function IDs from DT [0.00] PERCPU: Embedded 9 pages/cpu @ee7cc000 s13248 r8192 d15424 u36864 [0.00] Built 1 zonelists in Zone order, mobility grouping on. Total pages: 522768 [0.00] Kernel command line: console=ttyS0,115200 console=ttyS0,115200 [0.00] PID hash table entries: 4096 (order: 2, 16384 bytes) [0.00] Dentry cache hash table entries: 131072 (order: 7, 524288 bytes) [0.00] Inode-cache hash table entries: 65536 (order: 6, 262144 bytes) [0.00] Memory: 2057220K/2097152K available (6558K kernel code, 837K rwdata, 2208K rodata, 700K init, 396K bss, 39932K reserved, 1318912K highmem) [0.00] Virtual kernel memory layout: [0.00] vector : 0x - 0x1000 ( 4 kB) [0.00] fixmap : 0xffc0 - 0xffe0 (2048 kB) [0.00] vmalloc : 0xf000 - 0xff00 ( 240 MB) [0.00] lowmem : 0xc000 - 0xef80 ( 760 MB) [0.00] pkmap : 0xbfe0 - 0xc000 ( 2 MB) [0.00] modules : 0xbf00 - 0xbfe0 ( 14 MB) [0.00] .text : 0xc0008000 - 0xc0897da0 (8768 kB) [0.00] .init : 0xc0898000 - 0xc09473c0 ( 701 kB) [0.00] .data : 0xc0948000 - 0xc0a197f8 ( 838 kB) [0.00].bss : 0xc0a197f8 - 0xc0a7cbf4 ( 397 kB) [
Bug#767042: [jessie daily 2014-10-27] [armhf] Installation report: LeMaker Banana Pi - problems with autoloading the realtek ethernet PHY driver module
Control: tag -1 patch On Wed, Nov 05, 2014 at 09:52:40PM +0100, Karsten Merker wrote: [Failing ethernet PHY detection in d-i on the BananaPi] Further experiments show that increasing the startup-delay-us value in the regulator definition seems to solve the issue. I'll do some further experiments to determine a value that is long enough for a reliable detection without being unnecessary long and discuss the issue with the upstream author. A patch to solve the issue has been accepted upstream (see https://lists.infradead.org/pipermail/linux-arm-kernel/2014-November/301727.html). Attached is a backport of this patch for inclusion into the upcoming linux 3.16.7-3 package. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. Index: debian/patches/features/arm/dts-sunxi-Banana-Pi-increase-startup-delay-for-the-GMAC-PHY-regulator.patch === --- debian/patches/features/arm/dts-sunxi-Banana-Pi-increase-startup-delay-for-the-GMAC-PHY-regulator.patch (Revision 0) +++ debian/patches/features/arm/dts-sunxi-Banana-Pi-increase-startup-delay-for-the-GMAC-PHY-regulator.patch (Arbeitskopie) @@ -0,0 +1,35 @@ +From f82f99afaa65fd28d0f8409c50e8fcc65ee5e15b Mon Sep 17 00:00:00 2001 +From: Karsten Merker mer...@debian.org +Date: Wed, 12 Nov 2014 00:01:46 +0100 +Subject: ARM: dts: sunxi: Banana Pi: increase startup-delay for the GMAC PHY regulator +Origin: https://git.kernel.org/cgit/linux/kernel/git/mripard/linux.git/commit/?h=sunxi/dt-for-3.19id=f82f99afaa65fd28d0f8409c50e8fcc65ee5e15b + +On the LeMaker Banana Pi, probing the external ethernet PHY connected +to the SoC's internal GMAC module sometimes fails. The PHY power +supply is handled via a GPIO-controlled regulator, and the existing +regulator startup-delay of 5us is too short to make sure that the +PHY is always fully powered up when it is queried by phylib. Tests +have shown that to provide a reliable PHY detection, the startup-delay +has to be increased to at least 6us. To have a certain safety margin +and to cater for manufacturing variations between different boards, +the delay gets set to 10us as discussed on the linux-arm-kernel +mailinglist. + +Signed-off-by: Karsten Merker mer...@debian.org +Acked-by: Hans de Goede hdego...@redhat.com +Signed-off-by: Maxime Ripard maxime.rip...@free-electrons.com + +diff --git a/arch/arm/boot/dts/sun7i-a20-bananapi.dts b/arch/arm/boot/dts/sun7i-a20-bananapi.dts +index 3de847d..1cf1214 100644 +--- a/arch/arm/boot/dts/sun7i-a20-bananapi.dts b/arch/arm/boot/dts/sun7i-a20-bananapi.dts +@@ -207,7 +207,7 @@ + regulator-name = gmac-3v3; + regulator-min-microvolt = 330; + regulator-max-microvolt = 330; +- startup-delay-us = 5; ++ startup-delay-us = 10; + enable-active-high; + gpio = pio 7 23 0; + }; + Index: debian/patches/series === --- debian/patches/series (Revision 22061) +++ debian/patches/series (Arbeitskopie) @@ -113,6 +113,7 @@ features/arm/dts-sun7i-Add-spi0_pins_a-pinctrl-setting.patch features/arm/dts-sun7i-Add-uart3_pins_b-pinctrl-setting.patch features/arm/dts-sun7i-Add-Banana-Pi-board.patch +features/arm/dts-sunxi-Banana-Pi-increase-startup-delay-for-the-GMAC-PHY-regulator.patch features/arm/dts-sun7i-Add-support-for-Olimex-A20-OLinuXino-LIME.patch features/arm64/drivers-net-Add-APM-X-Gene-SoC-ethernet-driver-suppo.patch features/arm64/drivers-net-NET_XGENE-should-depend-on-HAS_DMA.patch
Bug#767042: [jessie daily 2014-10-27] [armhf] Installation report: LeMaker Banana Pi - problems with autoloading the realtek ethernet PHY driver module
On Wed, Nov 05, 2014 at 10:45:12AM +, Ian Campbell wrote: On Tue, 2014-11-04 at 22:37 +0100, Karsten Merker wrote: I have run further installation tests with today's current d-i images (still based on the same 3.16.5-1 kernel) OOI if you bodge your way through the install does the resulting system boot and discover the PHY reliably? IOW is it specific to d-i or not? Ethernet works in the installed system (tested with several cold and warm boots): [2.448442] stmmaceth 1c5.ethernet: no reset control found [2.454322] Ring mode enabled [2.457396] No HW DMA feature register supported [2.461941] Normal descriptors [2.465279] TX Checksum insertion supported [2.495563] libphy: stmmac: probed [2.499078] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active [2.505490] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) i.e. the PHY appears to have a seperate regulator on the BananaPi but not on the Cubietruck and I wonder whether the startup-delay-us = 5; might play a role here. I think that's a decent theory. Decent enoguh that it is probably worth taking it up with the sunxi kernel folks. Might also be the power supply differs between the two boards? Running the BananaPi with the Cubietruck's power supply does not change the behaviour. I have now run several tests with a modified BananaPi dtb in which I have added a regulator-always-on stanza to the reg_gmac_3v3 definition. With this change the PHY detection in d-i has worked every time, so this would support the theory that the regulator might not be powered up fast enough for the PHY detection to succeed, but I cannot see why this problem only occurs within the d-i environment. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#767042: [jessie daily 2014-10-27] [armhf] Installation report: LeMaker Banana Pi - problems with autoloading the realtek ethernet PHY driver module
On Wed, Nov 05, 2014 at 09:01:08PM +0100, Karsten Merker wrote: [Failing ethernet PHY detection in d-i on the BananaPi] I have now run several tests with a modified BananaPi dtb in which I have added a regulator-always-on stanza to the reg_gmac_3v3 definition. With this change the PHY detection in d-i has worked every time, so this would support the theory that the regulator might not be powered up fast enough for the PHY detection to succeed, but I cannot see why this problem only occurs within the d-i environment. Further experiments show that increasing the startup-delay-us value in the regulator definition seems to solve the issue. I'll do some further experiments to determine a value that is long enough for a reliable detection without being unnecessary long and discuss the issue with the upstream author. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#767042: [jessie daily 2014-10-27] [armhf] Installation report: LeMaker Banana Pi - problems with autoloading the realtek ethernet PHY driver module
On Mon, Nov 03, 2014 at 12:26:33AM +, Ben Hutchings wrote: On Sun, 2014-11-02 at 21:43 +0100, Karsten Merker wrote: [...] Following is the log from a d-i run where /sbin/modprobe and /sbin/insmod have been replaced by a script that logs all invocations of these two binaries with a modules-debug prefix before executing the original binary: [...] The entry at 00:03:37 does not show a modprobe invocation before the initialization of the stmmac module, so the module does not seem to be loaded by calling modprobe, which would explain why no autoloading of the PHY driver happens there. What I do not understand is that I also see no insmod invocation, and logging insmod invocations works when I run insmod from a shell. So the question remains: how is the stmmac module loaded at that point? I first thought that something might invoke kmod directly (i.e. not by calling the insmod/modprobe aliases), but logging direct kmod invocations also shows nothing, so I am probably overlooking something. Any ideas? So far as I can see, ethdetect runs hw-detect, which runs update-dev, which runs 'udevadm trigger --action=add --subsystem-nomatch=sound' and that should result in udev loading stmmac. (udev won't have done that when it started because the stmmac module wasn't included in the initramfs.) udev is now linked with libkmod, so it can load modules without running modprobe, and most driver modules get loaded that way. But as I said before, this isn't the case for the PHY driver modules. So, I think this shows the kernel is not running modprobe for some reason. Things are getting even stranger: I have run further installation tests with today's current d-i images (still based on the same 3.16.5-1 kernel) - http://d-i.debian.org/daily-images/armhf/daily/hd-media/hd-media.tar.gz dated 04-Nov-2014 05:16 and - http://cdimage.debian.org/cdimage/weekly-builds/armhf/jigdo-cd/debian-testing-armhf-CD-1.jigdo dated 2014-11-03 12:14) on a Cubietruck and on the BananaPi. On the Cubietruck everything works while on the BananaPi the PHY is not found. Both are booted from the same USB stick, so there cannot be any differences in the installer image. The hardware on both is rather similar (Allwinner A20 SoC with integrated MAC, external Realtek PHY) and they show the same PHY ID once the probing was successful, so I assume that they use identical PHY chips: Cubietruck: --- Jan 1 00:06:47 main-menu[179]: INFO: Menu item 'ethdetect' selected Jan 1 00:06:48 kernel: [ 78.983839] cfg80211: Calling CRDA to update world regulatory domain Jan 1 00:06:49 kernel: [ 78.998759] brcmfmac_sdio mmc1:0001:1: firmware: failed to load brcm/brcmfmac43362-sdio.bin (-2) Jan 1 00:06:49 kernel: [ 79.057067] stmmaceth 1c5.ethernet: no regulator found Jan 1 00:06:49 kernel: [ 79.057202] stmmaceth 1c5.ethernet: no reset control found Jan 1 00:06:49 kernel: [ 79.057218] Ring mode enabled Jan 1 00:06:49 kernel: [ 79.057226] No HW DMA feature register supported Jan 1 00:06:49 kernel: [ 79.057233] Normal descriptors Jan 1 00:06:49 kernel: [ 79.057241] TX Checksum insertion supported Jan 1 00:06:49 kernel: [ 79.171058] libphy: stmmac: probed Jan 1 00:06:49 kernel: [ 79.171089] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active Jan 1 00:06:49 kernel: [ 79.171100] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) BananaPi: - No Ethernet card was detected. If you know the name of the driver needed by your Ethernet card, you can select it from the list. -- shell ~ # lsmod |grep realtek ~ # dmesg | tail -8 [ 32.138512] ISO 9660 Extensions: RRIP_1991A [ 77.235117] stmmaceth 1c5.ethernet: no reset control found [ 77.235192] Ring mode enabled [ 77.235202] No HW DMA feature register supported [ 77.235211] Normal descriptors [ 77.235219] TX Checksum insertion supported [ 77.242678] libphy: stmmac: probed [ 77.242708] eth0: No PHY found ~ # rmmod stmmac ~ # modprobe stmmac ~ # lsmod |grep realtek realtek 1563 0 ~ # dmesg | tail -8 [ 148.823456] stmmaceth 1c5.ethernet: no reset control found [ 148.823487] Ring mode enabled [ 148.823496] No HW DMA feature register supported [ 148.823502] Normal descriptors [ 148.823509] TX Checksum insertion supported [ 148.854654] libphy: stmmac: probed [ 148.854683] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active [ 148.854694] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) ~ # The main visible difference between the two systems is that the Cubietruck also has a Broadcom 43362 SDIO Wifi module and d-i asks for the non-free firmware for it before the stmmac gets initialized, but I do not see how that should make a difference regarding the realtek PHY driver. What makes me wonder is the line stmmaceth 1c5.ethernet: no regulator found on the Cubietruck that does not appear on the BananaPi. Could there perhaps be some
Bug#767042: [jessie daily 2014-10-27] [armhf] Installation report: LeMaker Banana Pi - problems with autoloading the realtek ethernet PHY driver module
On Tue, Oct 28, 2014 at 11:14:40PM +0100, Karsten Merker wrote: On Mon, Oct 27, 2014 at 11:54:47PM +, Ben Hutchings wrote: I don't understand this. Karsten Merker mer...@debian.org (2014-10-27): [...] [ 73.104782] libphy: stmmac: probed [ 73.104812] eth0: No PHY found i.e. the correct ethernet MAC driver (stmmac) gets loaded automatically, but the necessary PHY driver (realtek) does not. [...] [ 499.392561] libphy: stmmac: probed [ 499.392592] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active [ 499.392604] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) and the ethernet interface works. The kernel version used in this installer build is 3.16.5-1. $ modinfo -F alias realtek mdio:???111001100100100010101 mdio:???111001100100100010010 In hex those are 1cc915 and 1cc912. (The 11 most significant bits are unspecified.) So modprobe certainly should find this module when requested by phylib. I have retried the installation today and tried something I had not done yesterday: just rmmod and directly afterwards modprobe the stmmac module. This results in the realtek PHY module getting auto-loaded, so the modprobe mechanism seems to work correctly there, but the question remains why this does not happen upon the first loading of the stmmac module. A protocol from d-i: No Ethernet card was detected. If you know the name of the driver needed by your Ethernet card, you can select it from the list. -- start shell ~ # lsmod Module Size Used by stmmac 73442 0 nls_utf81170 2 nls_cp437 4767 1 loop 16202 2 isofs 31318 1 vfat9621 1 fat52693 1 vfat ext4 485433 0 crc16 1146 1 ext4 mbcache 8210 1 ext4 jbd2 88199 1 ext4 sg 20824 0 sd_mod 38535 2 crc_t10dif 1041 1 sd_mod crct10dif_common1159 1 crc_t10dif usb_storage41751 1 ahci_sunxi 2652 0 libahci_platform4679 1 ahci_sunxi libahci23069 1 libahci_platform libata161761 3 libahci,libahci_platform,ahci_sunxi ohci_platform 4062 0 ohci_hcd 37591 1 ohci_platform scsi_mod 175644 4 sg,usb_storage,libata,sd_mod ehci_platform 4526 0 phy_sun4i_usb 4216 4 ehci_hcd 64373 1 ehci_platform sunxi_mmc 10557 0 ~ # dmesg | tail -8 [ 31.558145] ISO 9660 Extensions: RRIP_1991A [ 77.839161] stmmaceth 1c5.ethernet: no reset control found [ 77.839194] Ring mode enabled [ 77.839202] No HW DMA feature register supported [ 77.839210] Normal descriptors [ 77.839217] TX Checksum insertion supported [ 77.844560] libphy: stmmac: probed [ 77.844589] eth0: No PHY found ~ # rmmod stmmac ~ # modprobe stmmac ~ # dmesg | tail -8 [ 330.112850] stmmaceth 1c5.ethernet: no reset control found [ 330.112883] Ring mode enabled [ 330.112891] No HW DMA feature register supported [ 330.112898] Normal descriptors [ 330.112905] TX Checksum insertion supported [ 330.140101] libphy: stmmac: probed [ 330.140139] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active [ 330.140150] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) ~ # lsmod Module Size Used by realtek 1563 0 stmmac 73442 0 nls_utf81170 2 nls_cp437 4767 1 loop 16202 2 isofs 31318 1 vfat9621 1 fat52693 1 vfat ext4 485433 0 crc16 1146 1 ext4 mbcache 8210 1 ext4 jbd2 88199 1 ext4 sg 20824 0 sd_mod 38535 2 crc_t10dif 1041 1 sd_mod crct10dif_common1159 1 crc_t10dif usb_storage41751 1 ahci_sunxi 2652 0 libahci_platform4679 1 ahci_sunxi libahci23069 1 libahci_platform libata161761 3 libahci,libahci_platform,ahci_sunxi ohci_platform 4062 0 ohci_hcd 37591 1 ohci_platform scsi_mod 175644 4 sg,usb_storage,libata,sd_mod ehci_platform 4526 0 phy_sun4i_usb 4216 4 ehci_hcd 64373 1 ehci_platform sunxi_mmc 10557 0 As udev is *not* involved in loading MDIO PHY drivers (NIC drivers expect them to be bound synchronously) it isn't easy to monitor what's going on. You could replace modprobe with a script that logs its arguments to a file before calling the real modprobe. That should tell us whether the bug is in the kernel or userland. I
Bug#767042: [jessie daily 2014-10-27] [armhf] Installation report: LeMaker Banana Pi - problems with autoloading the realtek ethernet PHY driver module
On Mon, Oct 27, 2014 at 11:54:47PM +, Ben Hutchings wrote: On Tue, 2014-10-28 at 00:18 +0100, Cyril Brulebois wrote: Cc+=debian-kernel@ for input since I seem to recall having seen PHY drivers (including in a realtek context) being mentioned lately, at least on IRC, maybe on list as well. I don't understand this. Karsten Merker mer...@debian.org (2014-10-27): [...] [ 73.104782] libphy: stmmac: probed [ 73.104812] eth0: No PHY found i.e. the correct ethernet MAC driver (stmmac) gets loaded automatically, but the necessary PHY driver (realtek) does not. [...] [ 499.392561] libphy: stmmac: probed [ 499.392592] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active [ 499.392604] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) and the ethernet interface works. The kernel version used in this installer build is 3.16.5-1. $ modinfo -F alias realtek mdio:???111001100100100010101 mdio:???111001100100100010010 In hex those are 1cc915 and 1cc912. (The 11 most significant bits are unspecified.) So modprobe certainly should find this module when requested by phylib. I have retried the installation today and tried something I had not done yesterday: just rmmod and directly afterwards modprobe the stmmac module. This results in the realtek PHY module getting auto-loaded, so the modprobe mechanism seems to work correctly there, but the question remains why this does not happen upon the first loading of the stmmac module. A protocol from d-i: No Ethernet card was detected. If you know the name of the driver needed by your Ethernet card, you can select it from the list. -- start shell ~ # lsmod Module Size Used by stmmac 73442 0 nls_utf81170 2 nls_cp437 4767 1 loop 16202 2 isofs 31318 1 vfat9621 1 fat52693 1 vfat ext4 485433 0 crc16 1146 1 ext4 mbcache 8210 1 ext4 jbd2 88199 1 ext4 sg 20824 0 sd_mod 38535 2 crc_t10dif 1041 1 sd_mod crct10dif_common1159 1 crc_t10dif usb_storage41751 1 ahci_sunxi 2652 0 libahci_platform4679 1 ahci_sunxi libahci23069 1 libahci_platform libata161761 3 libahci,libahci_platform,ahci_sunxi ohci_platform 4062 0 ohci_hcd 37591 1 ohci_platform scsi_mod 175644 4 sg,usb_storage,libata,sd_mod ehci_platform 4526 0 phy_sun4i_usb 4216 4 ehci_hcd 64373 1 ehci_platform sunxi_mmc 10557 0 ~ # dmesg | tail -8 [ 31.558145] ISO 9660 Extensions: RRIP_1991A [ 77.839161] stmmaceth 1c5.ethernet: no reset control found [ 77.839194] Ring mode enabled [ 77.839202] No HW DMA feature register supported [ 77.839210] Normal descriptors [ 77.839217] TX Checksum insertion supported [ 77.844560] libphy: stmmac: probed [ 77.844589] eth0: No PHY found ~ # rmmod stmmac ~ # modprobe stmmac ~ # dmesg | tail -8 [ 330.112850] stmmaceth 1c5.ethernet: no reset control found [ 330.112883] Ring mode enabled [ 330.112891] No HW DMA feature register supported [ 330.112898] Normal descriptors [ 330.112905] TX Checksum insertion supported [ 330.140101] libphy: stmmac: probed [ 330.140139] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active [ 330.140150] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) ~ # lsmod Module Size Used by realtek 1563 0 stmmac 73442 0 nls_utf81170 2 nls_cp437 4767 1 loop 16202 2 isofs 31318 1 vfat9621 1 fat52693 1 vfat ext4 485433 0 crc16 1146 1 ext4 mbcache 8210 1 ext4 jbd2 88199 1 ext4 sg 20824 0 sd_mod 38535 2 crc_t10dif 1041 1 sd_mod crct10dif_common1159 1 crc_t10dif usb_storage41751 1 ahci_sunxi 2652 0 libahci_platform4679 1 ahci_sunxi libahci23069 1 libahci_platform libata161761 3 libahci,libahci_platform,ahci_sunxi ohci_platform 4062 0 ohci_hcd 37591 1 ohci_platform scsi_mod 175644 4 sg,usb_storage,libata,sd_mod ehci_platform 4526 0 phy_sun4i_usb 4216 4 ehci_hcd 64373 1 ehci_platform sunxi_mmc 10557 0 As udev is *not* involved in loading MDIO PHY drivers (NIC drivers expect them to be bound synchronously) it isn't easy to monitor what's going on. You could replace modprobe with a script that logs its arguments to a file before calling the real modprobe. That should tell us
Bug#767042: [jessie daily 2014-10-27] [armhf] Installation report: LeMaker Banana Pi - problems with autoloading the realtek ethernet PHY driver module
Package: installation-reports Boot method: hd-media (installer booted from USB stick) Image version: http://d-i.debian.org/daily-images/armhf/daily/hd-media/hd-media.tar.gz (dated 27-Oct-2014 05:17) together with http://cdimage.debian.org/cdimage/weekly-builds/armhf/jigdo-cd/debian-testing-armhf-CD-1.jigdo (dated 2014-10-27 11:58) Date: 2014-10-27 U-Boot: 2014.10+dfsg1-1 Machine: LeMaker Banana Pi Processor: Allwinner A20 (2x Cortex A7) Memory: 1GB Mass Storage: 16GB SD card Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[E] (see comments below, getting this to work requires manual loading of the proper ethernet PHY driver module) Configure network: [O] Detect CD: [O] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [O] Install base system:[O] Clock/timezone setup: [O] User/password setup:[O] Install tasks: [O] Install boot loader:[O] Overall install:[O] Comments/Problems: The detect network hardware step fails with No Ethernet card was detected. If you know the name of the driver needed by your Ethernet card, you can select it from the list. dmesg shows: [ 73.100072] stmmaceth 1c5.ethernet: no reset control found [ 73.100103] Ring mode enabled [ 73.100112] No HW DMA feature register supported [ 73.100120] Normal descriptors [ 73.100127] TX Checksum insertion supported [ 73.104782] libphy: stmmac: probed [ 73.104812] eth0: No PHY found i.e. the correct ethernet MAC driver (stmmac) gets loaded automatically, but the necessary PHY driver (realtek) does not. Loading the realtek module after the stmmac driver has probed for a PHY does not change anything; the stmmac driver does not re-probe for a PHY upon later loading of the PHY driver. Manually unloading the stmmac module, loading the PHY driver and then loading the stmmac module again from a shell by running ~ # rmmod stmmac ~ # modprobe realtek ~ # modprobe stmmac results in [ 499.364044] stmmaceth 1c5.ethernet: no reset control found [ 499.364076] Ring mode enabled [ 499.364084] No HW DMA feature register supported [ 499.364090] Normal descriptors [ 499.364097] TX Checksum insertion supported [ 499.392561] libphy: stmmac: probed [ 499.392592] eth0: PHY ID 001cc915 at 0 IRQ POLL (stmmac-0:00) active [ 499.392604] eth0: PHY ID 001cc915 at 1 IRQ POLL (stmmac-0:01) and the ethernet interface works. The kernel version used in this installer build is 3.16.5-1. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#764967: Please backport dts for the Olimex A20-OLinuXino-LIME
On Sun, 12 Oct 2014 18:23:24 +0200 Christian Kastner deb...@kvr.at wrote: would it be possible to include the dts for the Olimex A20-OLinuXino-LIME in 3.16 so that it can be used with Jessie? This device is almost identical to the already existing A10-OLinuXino-LIME; they only differ in the processor. The relevant commit from 3.17 is: a71b443 ARM: sun7i: Add support for Olimex A20-OLinuXino-LIME The patch from https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/patch/?id=a71b4438af8242f383906071205db95a8b8e7b6d does not apply cleanly against the current 3.16.5-2 kernel package SVN. A patch against current SVN with a refreshed version of the original patch (no content changes) is attached. A test build is currently running, but will take until tomorrow. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. Index: debian/patches/features/arm/dts-sun7i-Add-support-for-Olimex-A20-OLinuXino-LIME.patch === --- debian/patches/features/arm/dts-sun7i-Add-support-for-Olimex-A20-OLinuXino-LIME.patch (Revision 0) +++ debian/patches/features/arm/dts-sun7i-Add-support-for-Olimex-A20-OLinuXino-LIME.patch (Arbeitskopie) @@ -0,0 +1,160 @@ +From a71b4438af8242f383906071205db95a8b8e7b6d Mon Sep 17 00:00:00 2001 +From: FUKAUMI Naoki nao...@gmail.com +Date: Wed, 20 Aug 2014 14:25:03 +0900 +Subject: ARM: sun7i: Add support for Olimex A20-OLinuXino-LIME + +This patch adds support for Olimex A20-OLinuXino-LIME board. + +Signed-off-by: FUKAUMI Naoki nao...@gmail.com +Signed-off-by: Maxime Ripard maxime.rip...@free-electrons.com + +--- a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile +@@ -380,6 +380,7 @@ + sun7i-a20-cubieboard2.dtb \ + sun7i-a20-cubietruck.dtb \ + sun7i-a20-i12-tvbox.dtb \ ++ sun7i-a20-olinuxino-lime.dtb \ + sun7i-a20-olinuxino-micro.dtb + dtb-$(CONFIG_ARCH_TEGRA) += tegra20-harmony.dtb \ + tegra20-iris-512.dtb \ +--- /dev/null b/arch/arm/boot/dts/sun7i-a20-olinuxino-lime.dts +@@ -0,0 +1,137 @@ ++/* ++ * This is based on sun4i-a10-olinuxino-lime.dts ++ * ++ * Copyright 2014 - Hans de Goede hdego...@redhat.com ++ * Copyright (c) 2014 FUKAUMI Naoki nao...@gmail.com ++ * ++ * The code contained herein is licensed under the GNU General Public ++ * License. You may obtain a copy of the GNU General Public License ++ * Version 2 or later at the following locations: ++ * ++ * http://www.opensource.org/licenses/gpl-license.html ++ * http://www.gnu.org/copyleft/gpl.html ++ */ ++ ++/dts-v1/; ++/include/ sun7i-a20.dtsi ++/include/ sunxi-common-regulators.dtsi ++ ++/ { ++ model = Olimex A20-OLinuXino-LIME; ++ compatible = olimex,a20-olinuxino-lime, allwinner,sun7i-a20; ++ ++ soc@01c0 { ++ mmc0: mmc@01c0f000 { ++ pinctrl-names = default; ++ pinctrl-0 = mmc0_pins_a, mmc0_cd_pin_reference_design; ++ vmmc-supply = reg_vcc3v3; ++ bus-width = 4; ++ cd-gpios = pio 7 1 0; /* PH1 */ ++ cd-inverted; ++ status = okay; ++ }; ++ ++ usbphy: phy@01c13400 { ++ usb1_vbus-supply = reg_usb1_vbus; ++ usb2_vbus-supply = reg_usb2_vbus; ++ status = okay; ++ }; ++ ++ ehci0: usb@01c14000 { ++ status = okay; ++ }; ++ ++ ohci0: usb@01c14400 { ++ status = okay; ++ }; ++ ++ ahci: sata@01c18000 { ++ target-supply = reg_ahci_5v; ++ status = okay; ++ }; ++ ++ ehci1: usb@01c1c000 { ++ status = okay; ++ }; ++ ++ ohci1: usb@01c1c400 { ++ status = okay; ++ }; ++ ++ pinctrl@01c20800 { ++ ahci_pwr_pin_olinuxinolime: ahci_pwr_pin@1 { ++allwinner,pins = PC3; ++allwinner,function = gpio_out; ++allwinner,drive = 0; ++allwinner,pull = 0; ++ }; ++ ++ led_pins_olinuxinolime: led_pins@0 { ++allwinner,pins = PH2; ++allwinner,function = gpio_out; ++allwinner,drive = 1; ++allwinner,pull = 0; ++ }; ++ }; ++ ++ uart0: serial@01c28000 { ++ pinctrl-names = default; ++ pinctrl-0 = uart0_pins_a; ++ status = okay; ++ }; ++ ++ i2c0: i2c@01c2ac00 { ++ pinctrl-names = default; ++ pinctrl-0 = i2c0_pins_a; ++ status = okay; ++ ++ axp209: pmic@34 { ++compatible = x-powers,axp209; ++reg = 0x34; ++interrupt-parent = nmi_intc; ++interrupts = 0 8; ++ ++interrupt-controller; ++#interrupt-cells = 1; ++ }; ++ }; ++ ++ gmac: ethernet@01c5 { ++ pinctrl-names = default; ++ pinctrl-0 = gmac_pins_mii_a; ++ phy = phy1; ++ phy-mode = mii; ++ status = okay; ++ ++ phy1: ethernet-phy@1 { ++reg = 1; ++ }; ++ }; ++ }; ++ ++ leds { ++ compatible = gpio-leds; ++ pinctrl-names = default; ++ pinctrl-0 = led_pins_olinuxinolime; ++ ++ green { ++ label = a20-olinuxino-lime:green:usr; ++ gpios = pio 7 2 0; ++ default-state = on; ++ }; ++ }; ++ ++ reg_ahci_5v: ahci-5v { ++ pinctrl-0 = ahci_pwr_pin_olinuxinolime; ++ gpio = pio 2 3 0; ++ status = okay; ++ }; ++ ++
Bug#763005: installation-guide is marked for autoremoval from testing
Control: reassign -1 www.debian.org Control: severity -1 normal thanks On Sun, Oct 12, 2014 at 04:39:03AM +, Debian testing autoremoval watch wrote: installation-guide 20140916 is marked for autoremoval from testing on 2014-10-26 It is affected by these RC bugs: 763005: installation-guide-amd64: website for wheezy leads to jessie installation guide From looking at the bug report this has not actually been a bug in the installation-guide package, but a problem in the cron job that puts the installtion guide onto the website, so this should not result in a package removal from testing. The cron job has been manually fixed up, so the actual problem (wrong version of the guide on the website) does not occur anymore at the moment. What is still open is how to change the cronjob to handle this better/automatically in the future, but that IMHO does not justify an RC bug against the manual itself. I therefore assign this bug to www.debian.org and lower the severity to normal. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#763897: linux 3.16.3-2: [PATCH] Please add a dts for the LeMaker Banana Pi
Source: linux Version: 3.16.3-2 Severity: Wishlist Tags: patch Hello, the LeMaker Banana Pi is a small ARM development board based on the Allwinner A20 SoC. Several DDs have received one of these boards at Debconf 14 and it would be really nice to have it supported in Jessie's kernel 3.16. Driver support for the board (UART, MMC, SATA, USB, Ethernet) is already available in kernel 3.16, just the device-tree information is missing. A dts patchset for the board has been written by Hans de Goede (see http://lists.infradead.org/pipermail/linux-arm-kernel/2014-October/292110.html) and has been accepted upstream (see http://lists.infradead.org/pipermail/linux-arm-kernel/2014-October/292431.html). Attached is a patch against the current sid linux package SVN tree that contains a backport of this patchset to kernel 3.16. No content changes were applied to the patchset, the patches were just refreshed to apply cleanly against 3.16. Included is also another dts patch from linux-next, on which the patchset depends: https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/?id=0fc2b7af8cd5918c0101dfb178b5a3a4b021a50b I would apprechiate very much if you could include this patch into the linux 3.16.3-3 package for Jessie. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. Index: debian/patches/features/arm/dts-sun7i-Add-Banana-Pi-board.patch === --- debian/patches/features/arm/dts-sun7i-Add-Banana-Pi-board.patch (revision 0) +++ debian/patches/features/arm/dts-sun7i-Add-Banana-Pi-board.patch (working copy) @@ -0,0 +1,244 @@ +From: Hans de Goede hdego...@redhat.com +Subject: [PATCH v2 3/3] ARM: dts: sun7i: Add Banana Pi board +Date: Wed, 1 Oct 2014 09:26:06 +0200 + +The Banana Pi is an A20 based development board using Raspberry Pi compatible +IO headers. It comes with 1 GB RAM, 1 Gb ethernet, 2x USB host, sata, hdmi +and stereo audio out + various expenansion headers: + +http://www.lemaker.org/ + +Signed-off-by: Hans de Goede hdego...@redhat.com +--- + arch/arm/boot/dts/Makefile | 1 + + arch/arm/boot/dts/sun7i-a20-bananapi.dts | 214 +++ + 2 files changed, 215 insertions(+) + create mode 100644 arch/arm/boot/dts/sun7i-a20-bananapi.dts + +--- a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile +@@ -376,6 +376,7 @@ + sun6i-a31-colombus.dtb \ + sun6i-a31-m9.dtb + dtb-$(CONFIG_MACH_SUN7I) += \ ++ sun7i-a20-bananapi.dtb \ + sun7i-a20-cubieboard2.dtb \ + sun7i-a20-cubietruck.dtb \ + sun7i-a20-i12-tvbox.dtb \ +--- /dev/null b/arch/arm/boot/dts/sun7i-a20-bananapi.dts +@@ -0,0 +1,214 @@ ++/* ++ * Copyright 2014 Hans de Goede hdego...@redhat.com ++ * ++ * Hans de Goede hdego...@redhat.com ++ * ++ * This file is dual-licensed: you can use it either under the terms ++ * of the GPL or the X11 license, at your option. Note that this dual ++ * licensing only applies to this file, and not this project as a ++ * whole. ++ * ++ * a) This library is free software; you can redistribute it and/or ++ * modify it under the terms of the GNU General Public License as ++ * published by the Free Software Foundation; either version 2 of the ++ * License, or (at your option) any later version. ++ * ++ * This library is distributed in the hope that it will be useful, ++ * but WITHOUT ANY WARRANTY; without even the implied warranty of ++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ++ * GNU General Public License for more details. ++ * ++ * You should have received a copy of the GNU General Public ++ * License along with this library; if not, write to the Free ++ * Software Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, ++ * MA 02110-1301 USA ++ * ++ * Or, alternatively, ++ * ++ * b) Permission is hereby granted, free of charge, to any person ++ * obtaining a copy of this software and associated documentation ++ * files (the Software), to deal in the Software without ++ * restriction, including without limitation the rights to use, ++ * copy, modify, merge, publish, distribute, sublicense, and/or ++ * sell copies of the Software, and to permit persons to whom the ++ * Software is furnished to do so, subject to the following ++ * conditions: ++ * ++ * The above copyright notice and this permission notice shall be ++ * included in all copies or substantial portions of the Software. ++ * ++ * THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY KIND, ++ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES ++ * OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND ++ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT ++ * HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, ++ *
Bug#762634: initramfs-tools: [armhf] mounting rootfs on USB disk fails / some USB host controller drivers missing in initramfs
On Tue, Sep 23, 2014 at 11:42:08PM +0200, Michael Prokop wrote: * Karsten Merker [Tue Sep 23, 2014 at 11:22:32PM +0200]: Please always include ohci-platform, ehci-platform and phy-sun4i-usb into the initramfs if they are provided by the kernel for which the initramfs is built. Does this look like it would provide what you're asking for? http://anonscm.debian.org/cgit/kernel/initramfs-tools.git/commit/?h=mika/bug_762634id=3b835665015c0a9287284c2548b12ab7c8cabc78 Hello, outside d-i and for the MODULES=most configuration, this works fine: $ grep MODULES /etc/initramfs-tools/initramfs.conf # MODULES: [ most | netboot | dep | list ] MODULES=most $ sudo update-initramfs -u -k 3.16-2-armmp-lpae update-initramfs: Generating /boot/initrd.img-3.16-2-armmp-lpae Installing sun7i-a20-cubietruck.dtb 3.16-2-armmp-lpae into /boot Installing sun7i-a20-cubietruck.dtb 3.16-2-armmp-lpae into /boot flash-kernel: installing version 3.16-2-armmp-lpae Generating boot script u-boot image... done. Taking backup of boot.scr. Installing new boot.scr. $ lsinitramfs /boot/initrd.img-3.16-2-armmp-lpae |grep -e hci-platform -e phy-sun4i-usb lib/modules/3.16-2-armmp-lpae/kernel/drivers/phy/phy-sun4i-usb.ko lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb/host/ohci-platform.ko lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb/host/ehci-platform.ko Inside the /target chroot created by d-i, update-initramfs is by default configured to run with MODULES=dep and gives only the following modules on my armhf/sunxi test system: lib/modules/3.16-2-armmp-lpae/kernel/drivers lib/modules/3.16-2-armmp-lpae/kernel/drivers/scsi lib/modules/3.16-2-armmp-lpae/kernel/drivers/scsi/scsi_mod.ko lib/modules/3.16-2-armmp-lpae/kernel/drivers/scsi/sd_mod.ko lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb/host lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb/host/ehci-hcd.ko lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb/host/ehci-platform.ko lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb/storage lib/modules/3.16-2-armmp-lpae/kernel/drivers/usb/storage/usb-storage.ko lib/modules/3.16-2-armmp-lpae/kernel/drivers/md lib/modules/3.16-2-armmp-lpae/kernel/drivers/md/dm-mod.ko lib/modules/3.16-2-armmp-lpae/kernel/lib lib/modules/3.16-2-armmp-lpae/kernel/lib/crc16.ko lib/modules/3.16-2-armmp-lpae/kernel/lib/crc-t10dif.ko lib/modules/3.16-2-armmp-lpae/kernel/fs lib/modules/3.16-2-armmp-lpae/kernel/fs/mbcache.ko lib/modules/3.16-2-armmp-lpae/kernel/fs/ext4 lib/modules/3.16-2-armmp-lpae/kernel/fs/ext4/ext4.ko lib/modules/3.16-2-armmp-lpae/kernel/fs/jbd2 lib/modules/3.16-2-armmp-lpae/kernel/fs/jbd2/jbd2.ko lib/modules/3.16-2-armmp-lpae/kernel/crypto lib/modules/3.16-2-armmp-lpae/kernel/crypto/crct10dif_generic.ko lib/modules/3.16-2-armmp-lpae/kernel/crypto/crct10dif_common.ko The phy-sun4i-usb module is missing there, although it is loaded by the running kernel: # lsmod |grep phy_sun4i_usb phy_sun4i_usb 4260 4 Without phy_sun4i_usb, USB support does not work at all on this platform, so this module would always have to be included. The missing ohci-hcd and ohci-platform modules would be explained by the fact that this particular device technically has both EHCI and OHCI controllers, but the platform-dependent OHCI glue code is not yet implemented in the kernel, so the OHCI part is currently invisible to the kernel. When manually running update-initramfs with MODULES=most inside the d-i /target chroot, all three modules (ohci-platform, ehci-platform and phy-sun4i-usb) are there, but by default the user is not presented with a choice regarding MODULES=dep vs. MODULES=most in d-i. The relevant debconf prompt is only presented at debconf priority medium, i.e. when running d-i in expert mode, so I would appreciate very much if you could include the missing phy-sun4i-usb module also when building the initramfs with MODULES=dep. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#762634: initramfs-tools: [armhf] mounting rootfs on USB disk fails / some USB host controller drivers missing in initramfs
On Thu, Sep 25, 2014 at 01:31:16PM +0200, Michael Prokop wrote: * Cyril Brulebois [Thu Sep 25, 2014 at 12:32:22PM +0200]: Michael Prokop m...@debian.org (2014-09-25): * Karsten Merker [Thu Sep 25, 2014 at 08:15:48AM +0200]: Inside the /target chroot created by d-i, update-initramfs is by default configured to run with MODULES=dep and gives only the following modules on my armhf/sunxi test system: Hmpf, I so much am not a fan of this default MODULES=dep behaviour of d-i... Can you please clarify? I see this in base-installer: | if db_get base-installer/initramfs-tools/driver-policy \ |[ -z $RET ]; then | # Get default for architecture [snip] Granted, no coffee yet. But I seem to recall that people having issues with MODULES=dep are those who actually selected it manually (e.g. through expert install). I'm just not a friend of MODULES=dep as a default behaviour, good to know that d-i uses a sane default here. :) Thanks for verifying! My local test confirms that a basic installation (netboot-gtk image on amd64 using udebs from sid) leads to MODULES=most in /target. I didn't go further than the popcon prompt though. Hello, I have just run another test on armhf with today's daily d-i build (dated 25-Sep-2014 05:17) and default debconf priority (at which no debconf prompt regarding initramfs-tools gets displayed). This results in: ~ # cat /target/etc/initramfs-tools/conf.d/driver-policy # Driver inclusion policy selected during installation # Note: this setting overrides the value set in the file # /etc/initramfs-tools/initramfs.conf MODULES=dep and the short module list in the initramfs confirms that it was really built with MODULES=dep. The comment marked above suggests that this is an architecture-dependent default, so we could possibly both be right ;-). If yes, we will have to find a solution regarding the inclusion of the phy-sun4i-usb module when building the initramfs with MODULES=dep, though. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#762634: initramfs-tools: [armhf] mounting rootfs on USB disk fails / some USB host controller drivers missing in initramfs
On Thu, Sep 25, 2014 at 09:21:57AM +0200, Michael Prokop wrote: Is there a /sys/... entry which would make it obvious for i-t that phy-sun4i-usb exists/should be loaded? Sorry, I do not know that. A possible approach could be parsing some of the information in /proc/device-tree, as the module initialization is triggered by the existence of certain compatible properties in the device-tree. From drivers/phy/phy-sun4i-usb.c in the kernel sources: static const struct of_device_id sun4i_usb_phy_of_match[] = { { .compatible = allwinner,sun4i-a10-usb-phy }, { .compatible = allwinner,sun5i-a13-usb-phy }, { .compatible = allwinner,sun6i-a31-usb-phy }, { .compatible = allwinner,sun7i-a20-usb-phy }, { }, }; MODULE_DEVICE_TABLE(of, sun4i_usb_phy_of_match); Something like grepping for the compatible strings in the result of a find /proc/device-tree/ -iname compatible might work, but I have not actually tested that yet. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#762634: [armhf] mounting rootfs on USB disk fails / some USB host controller drivers missing in initramfs
Package: initramfs-tools Version: 0.116 Hello, running Debian with the root filesystem on a USB mass storage device (such as a USB harddisk) requires that the driver modules for the USB host controllers of the system are available in the initramfs. If they are missing, the root filesystem cannot be mounted, which currently happens on a number of armhf systems. On i386/amd64, the OHCI/EHCI host controllers are PCI devices which are supported by the ohci-pci and ehci-pci modules. On many armhf systems the USB host controllers are OHCI/EHCI-compatible, but implemented as a platform device, so they are not supported by ohci-pci and ehci-pci. Instead these systems need the following platform device driver modules: - ohci-platform - ehci-platform and in the case of armhf/sunxi-based systems an additional USB phy driver module: - phy-sun4i-usb These modules are not included in the initramfs built by initramfs-tools (even if MODULES=most is set in initramfs.conf), which makes it impossible to boot with the rootfs on a USB disk on such systems. Please always include ohci-platform, ehci-platform and phy-sun4i-usb into the initramfs if they are provided by the kernel for which the initramfs is built. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#762253: Tasksel 3.25: fails due to wrong apt invocation (missing -o in front of additional apt option)
Package: tasksel Version: 3.25 Severity: grave Justification: breaks d-i Hello, the upload of tasksel 3.25 has broken tasksel in d-i: Sep 20 06:04:28 pkgsel: starting tasksel Sep 20 06:04:56 in-target: E Sep 20 06:04:56 in-target: : Sep 20 06:04:56 in-target: Invalid operation APT::Acquire::Retries=3 Sep 20 06:04:56 in-target: Sep 20 06:04:56 in-target: tasksel: apt-get failed (100) Sep 20 06:04:57 main-menu[190]: WARNING **: Configuring 'pkgsel' failed with error code 1 Sep 20 06:04:57 main-menu[190]: WARNING **: Menu item 'pkgsel' failed. This is caused by the following commit: http://anonscm.debian.org/cgit/tasksel/tasksel.git/commit/?id=645455083756a71f1843c3deebdb73bc6324c66a where a -o is missing before the added APT::Acquire::Retries=3. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#762219: linux 3.16.3-1 FTBS for the armel kirkwood flavor due to size limit
source: linux version: 3.16.3-1 [CCing debian-arm@lists.d.o] Hello, according to https://buildd.debian.org/status/fetch.php?pkg=linuxarch=armelver=3.16.3-1stamp=1411133746 linux 3.16.3-1 FTBFS for the armel kirkwood flavor due to the resulting image being too big for the platform by 648 bytes, so the kernel configuration for this platform will probably have to be trimmed down: make[3]: Leaving directory '/«PKGBUILDDIR»/debian/build/build_armel_none_kirkwood' python debian/bin/buildcheck.py debian/build/build_armel_none_kirkwood armel none kirkwood Can't read ABI reference. ABI not checked! Continuing. Image too large (2097728 2097080)! Refusing to continue. make[2]: *** [debian/stamps/build_armel_none_kirkwood_plain] Error 1 debian/rules.real:158: recipe for target 'debian/stamps/build_armel_none_kirkwood_plain' failed Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#762221: linux 3.16.3-1 FTFBS for s390x with undefined reference to compat_sys_ni_syscall
source: linux version: 3.16.3-1 [CCing debian-s390@lists.d.o] According to https://buildd.debian.org/status/fetch.php?pkg=linuxarch=s390xver=3.16.3-1stamp=1411128376 linux 3.16.3-1 FTBFS on s390x with the following error: LINKvmlinux LD vmlinux.o MODPOST vmlinux.o GEN .version CHK include/generated/compile.h UPD include/generated/compile.h CC init/version.o LD init/built-in.o arch/s390/built-in.o: In function `sys_call_table_emu': (.rodata+0x2bbc): undefined reference to `compat_sys_ni_syscall' arch/s390/built-in.o: In function `sys_call_table_emu': (.rodata+0x2bc0): undefined reference to `compat_sys_ni_syscall' make[5]: *** [vmlinux] Error 1 /«PKGBUILDDIR»/Makefile:891: recipe for target 'vmlinux' failed make[4]: *** [sub-make] Error 2 Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#761591: linux 3.16.2-3: missing usb platform modules in udeb
On Sun, Sep 14, 2014 at 10:31:56PM +0200, Karsten Merker wrote: the usb-modules udeb built by the linux 3.16.2-3 package does not contain the usb platform driver modules ehci-platform and ohci-platform. Systems that have EHCI- or OHCI-compatible USB host controllers which are not implemented as PCI devices, but as platform devices, need these modules to provide the necessary glue to make ehci-hcd and ohci-hcd work. Without them, there is no USB support in the Debian-installer on e.g. all sunxi-based devices. Attached is a small patch that should add the necessary modules to the udeb (test build still running). The test build has completed in the meantime and I have locally built a debian-installer based on it. Unfortunately my original patch has not been sufficient - besides the (ehci|ohci)-platform modules, on sunxi-based devices an additional phy driver (phy-sun4i-usb) appears to be required. Follwing is a new version of my original patch - a test build is running but due to the build host being rather slow, I probably won't be able to actually test it in d-i before tomorrow evening. Index: debian/installer/modules/usb-modules === --- debian/installer/modules/usb-modules(Revision 21845) +++ debian/installer/modules/usb-modules(Arbeitskopie) @@ -1,7 +1,10 @@ ehci-hcd ? ehci-pci ? +ehci-platform ? ohci-hcd ? ohci-pci ? +ohci-platform ? uhci-hcd ? xhci-hcd ? usbcore ? +phy-sun4i-usb ? Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#761514: [armhf/sunxi] [d-i daily 20140914] Cubietruck installation report (installation to MMC)
Package: installation-reports Boot method: netboot Image version: http://d-i.debian.org/daily-images/armhf/daily/netboot/ dated 2014-09-14 05:17 Install Date: 2014-09-14 Machine: Cubietech Cubietruck Processor: Allwinner A20 (2x Cortex A7) Memory:2GB U-boot:u-boot 2014.10~rc2+dfsg1-1 from experimental Console: serial console Partitions (SD card): /dev/mmcblk0p1 ext2240972 10175218356 5% /boot /dev/mmcblk0p2 ext4 14136820 802108 12593544 6% / Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[O] Configure network: [O] Detect CD: [ ] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [O] Install base system:[O] Clock/timezone setup: [O] User/password setup:[O] Install tasks: [O] Install boot loader:[O] Overall install:[O] Comments: - The installation was done in expert mode (debconf priority medium). - I selected Sid instead of Jessie as target suite as kernel 3.16 (required for installation on MMC) has not yet migrated to Jessie. - Partitioning was done using the guided partitioning option. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#761514: [armhf/sunxi] [d-i daily 20140914] Cubietruck installation report (installation to MMC)
On Sun, Sep 14, 2014 at 05:25:57PM +0200, Cyril Brulebois wrote: Karsten Merker mer...@debian.org (2014-09-14): U-boot:u-boot 2014.10~rc2+dfsg1-1 from experimental [...] Comments: - The installation was done in expert mode (debconf priority medium). - I selected Sid instead of Jessie as target suite as kernel 3.16 (required for installation on MMC) has not yet migrated to Jessie. - Partitioning was done using the guided partitioning option. I see you also need u-boot from experimental; I'm not sure it'll reach sid + testing in time for the next release. Are you in touch with its maintainer/do you his short term plans? [CCing Vagrant Cascadian (Debian u-boot package maintainer) and Ian Campbell (mainline u-boot sunxi platform custodian)] U-Boot v2014.10 is scheduled to be released on Oct 13, 2014. Vagrant is tracking the rc releases in experimental and the plan is to get the final u-boot v2014.10 into unstable as soon as it is released. That is a tight schedule, but I think it is feasible. The mainline u-boot version currently in jessie (2014.07+dfsg1-1) has only very limited sunxi platform support. It works on the Cubietruck, but does not support several other sunxi-based systems for which we have support in d-i, so we are very strongly interested in getting v2014.10 into Jessie. D-i can also work with older u-boot-sunxi versions that are supplied by some board vendors (flash-kernel takes explicit care in the boot scripts it creates to handle this properly), but those older versions do not have the full featureset that mainline u-boot v2014.10 provides. The two major features missing in the vendor-supplied versions are - PSCI support, without which there is no SMP, i.e. the kernel can only initalize the first CPU core - AHCI support, i.e. booting from SATA harddisk is not possible. So using old u-boot-sunxi versions instead of current mainline is technically possible, but from a user point of view really undesireable. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#751704: libparted ped_disk_clobber() overwrites firmware on some arm systems
Control: tags -1 pending thanks On Thu, Aug 28, 2014 at 10:31:43PM +0200, Karsten Merker wrote: as libparted upstream has now confirmed that modifying PedDisk.needs_clobber from within the calling application is ok, I would like to apply the following patch to partman-base unless somebody has further objections against it. Functionally it is the same patch that I had posted earlier already, just with some coding style cleanups. The patch has been committed to the partman-base git repository. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#751704: libparted ped_disk_clobber() overwrites firmware on some arm systems
Hello, as libparted upstream has now confirmed that modifying PedDisk.needs_clobber from within the calling application is ok, I would like to apply the following patch to partman-base unless somebody has further objections against it. Functionally it is the same patch that I had posted earlier already, just with some coding style cleanups. Regards, Karsten From bd3e4f79ea620fefb48c768dec22a3f7d1ad31a2 Mon Sep 17 00:00:00 2001 From: Karsten Merker mer...@debian.org Date: Thu, 28 Aug 2014 20:38:12 +0200 Subject: [PATCH] Take care of the firmware area on sunxi-based systems By default partman calls ped_disk_clobber when writing a new partition table, but on the MMC device of sunxi-based systems this would overwrite the firmware area, resulting in an unbootable system (see bug #751704). Handle this as a special case in command_commit(). --- parted_server.c | 33 + 1 file changed, 33 insertions(+) diff --git a/parted_server.c b/parted_server.c index 55cf151..a6283f6 100644 --- a/parted_server.c +++ b/parted_server.c @@ -1330,6 +1330,25 @@ command_dump() oprintf(OK\n); } +/* Check whether we are running on a sunxi-based system. */ +int +is_sunxi_system() +{ +int cpuinfo_handle; +int result = 0; +char buf[4096]; +int length; + +if ((cpuinfo_handle = open(/proc/cpuinfo, O_RDONLY)) != -1) { +length = read(cpuinfo_handle, buf, sizeof(buf)-1); +buf[length]='\0'; +if (strstr(buf, Allwinner) != NULL) +result = 1; +close(cpuinfo_handle); +} +return result; +} + void command_commit() { @@ -1337,6 +1356,20 @@ command_commit() if (dev == NULL) critical_error(The device %s is not opened., device_name); log(command_commit()); + +/* The boot device on sunxi-based systems needs special handling. + * By default partman calls ped_disk_clobber when writing the + * partition table, but on sunxi-based systems this would overwrite + * the firmware area, resulting in an unbootable system (see + * bug #751704). + */ +if (is_sunxi_system() !strcmp(disk-dev-path, /dev/mmcblk0)) { +disk-needs_clobber = 0; +log(Sunxi platform detected. Disabling ped_disk_clobber \ +for the boot device %s to protect the firmware \ +area., disk-dev-path); +} + open_out(); if (disk != NULL named_is_changed(device_name)) ped_disk_commit(disk); -- 2.1.0 -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#751704: libparted ped_disk_clobber() overwrites firmware on some arm systems
Hello, the following is a discussion from the Debian bugtracking system regarding Debian Bug#751704 (https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=751704). It needs involvement from parted upstream, therefore I am forwarding it to bug-par...@gnu.org. Kind regards, Karsten - Forwarded message from Jim Meyering j...@meyering.net - Date: Sun, 17 Aug 2014 14:51:41 -0700 From: Jim Meyering j...@meyering.net Subject: Bug#751704: libparted questions (was: Bug#751704: partman-base 173: partman overwrites parts of u-boot) On Wed, Aug 13, 2014 at 12:07 PM, Karsten Merker mer...@debian.org wrote: [CCing Otavio Salvador and Jim Meyering; the following is a short summary of the situation; the full history can be read in bug #751704: Debian-Installer uses partman for partitioning, which in turn is based on libparted. On sunxi-based systems, upon writing the partition table, partman/libparted overwrites parts of u-boot which are located between the end of the partition table and the beginning of the first partition which results in an unbootable system. An attempt to solve this problem has been to conditionally set PedDisk.needs_clobber to 0 in partman when it detects that it is trying to write to a boot device on sunxi-based systems.] Colin Watson cjwat...@debian.org wrote: PedDisk.needs_clobber is marked as office use only in parted; I interpret that as meaning that it really shouldn't be fiddled with outside parted, even though it's technically exposed. Could you please look at fixing parted to avoid clobbering the firmware area instead? I think that would be more correct. I have no idea what is really meant with the comment /* office use only ;-) */ int needs_clobber; in /usr/include/parted/disk.h, so I would like to ask upstream for clarification. Otavio, Jim: you are listed as parted upstream maintainers on http://www.gnu.org/software/parted/ - could you shed some light on this? Is it ok for an application to fiddle with the needs_clobber element or is this to be considered strictly libparted-internal? @Colin: If the solution is to be completely encapsulated in libparted, I see a large problem in how to solve the conflict between space after the partition table being very differently used by firmware for different SoCs on one side, and libparted trying to make sure that there are no remains of other partition table formats in the respective area on the other side - at least with the prerequisite of keeping both the current defaults (clobbering) as well as keeping the libparted API unchanged. With the current there is one erase size for all platforms method in ped_disk_clobber() in libparted/disk.c, we inevitably end up with conflicting requirements. An example: the source for ped_disk_clobber() states: /* How many sectors to zero out at each end. This must be large enough to zero out the magic bytes starting at offset 8KiB on a DASD partition table. Doing the same from the end of the disk is probably overkill, but at least on GPT, we do need to zero out the final sector. */ So for at least one platform (according to Wikipedia DASD seems to be some s/390 format), the area at offset 8KiB has to be wiped out while for another (armhf/sunxi) it may not be wiped out as the u-boot SPL is located there and cannot be relocated because its sector address is hardcoded in the SoC. According to https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=751704#25, similar problems (but at other offsets) come up for other SoCs. According to the examples listed there, for TI SoCs libparted would have to stop clobbering the GPT header after writing a DOS MBR, but could wipe the DASD area. For a Freescale i.MX SoC libparted could legally clobber the GPT header, but would have to refrain from clobbering the areas behind it. If you extrapolate this to the large number of different SoC families, to handle this completely inside libparted, the library would have to know what exactly is the target system for which it writes a partition table and special-case the handling for each of them. While one might assume that the partition table is for an s/390 system when a DASD partition table is generated, this does not work as easily for the plethora of different architectures and systems that use DOS MBR partition tables. This gets even more complicated by the fact that libparted may run on one platform but modify partition tables for another platform, such as when operating on disk images for use with emulators or when operating on hd-media images for different arm platforms (with different firmware/bootloader requirements) on one autobuild host, so trying to do runtime detection of the host system would still not cover all use cases. In the case of creating images from scratch, the problem can be circumvented by (re-)writing
Bug#751704: bug#18289: libparted ped_disk_clobber() overwrites firmware on some arm systems
On Mon, Aug 18, 2014 at 09:59:49AM -0700, Brian C. Lane wrote: On Mon, Aug 18, 2014 at 08:19:17AM +0200, Karsten Merker wrote: Hello, the following is a discussion from the Debian bugtracking system regarding Debian Bug#751704 (https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=751704). It needs involvement from parted upstream, therefore I am forwarding it to bug-par...@gnu.org. Thanks for forwarding this. parted should only be clobbering these extra sectors when a new disklabel is applied (eg. mklabel in the ui) which, I think, is the appropriate thing to do. If you are operating on an existing disklabel and want to preserver a boot loader in the space before the 1st partition you shouldn't be calling ped_disk_new_fresh(). If you are creating a new disklabel then any boot loader code should be installed after partitioning. I don't think parted needs any changes. Hello, thanks for your swift reply. I fully understand your position, but unfortunately things on arm are fundamentally different from the PC world. U-Boot is more of a BIOS than a bootloader like GRUB; it is highly board specific and handles low-level stuff such as setting the IO pinmuxing for the specific board and configuring the DRAM controller. Without it, the board is completely dead from a user perspective. On a PC, the BIOS is always available in ROM/Flash even when all disk devices have been wiped and the user can still select some other device (such as the network, a CDROM drive or a USB memory stick) to boot from. On the arm systems we are talking about, there is no ROM BIOS and the u-boot image on the SD card (or on an eMMC chip) is the only way to interact with the system at all. The usual case is that the u-boot image is written raw onto the storage medium, i.e. there is no partition table or filesystem on it by default, so we need to create a new disklabel in the installer. You are fully right that normally a bootloader should be installed after partitioning. This works well for the case of a bootloader that uses universally available BIOS functions and is not hardware-specific, such as is the case on PCs. In the case of u-boot on the other hand, in PC-lingo we would have to provide the installer with the ability to flash the BIOS for every PC model on the market as part of the operating system installation, which is not feasible. We might be able to do that for a small selection of devices, but not for all of them -- we need to keep the u-boot image that is on the device even when creating a new disklabel, but we are unsure what is the best way to handle this. The approach in https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=751704#60 (setting PedDisk.needs_clobber to 0 before calling ped_disk_commit for specific devices) works in practice, but the question was whether it is ok for the calling application to fiddle around with the needs_clobber flag, or whether we should take some other approach. Kind Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#751704: libparted questions (was: Bug#751704: partman-base 173: partman overwrites parts of u-boot)
[CCing Otavio Salvador and Jim Meyering; the following is a short summary of the situation; the full history can be read in bug #751704: Debian-Installer uses partman for partitioning, which in turn is based on libparted. On sunxi-based systems, upon writing the partition table, partman/libparted overwrites parts of u-boot which are located between the end of the partition table and the beginning of the first partition which results in an unbootable system. An attempt to solve this problem has been to conditionally set PedDisk.needs_clobber to 0 in partman when it detects that it is trying to write to a boot device on sunxi-based systems.] Colin Watson cjwat...@debian.org wrote: PedDisk.needs_clobber is marked as office use only in parted; I interpret that as meaning that it really shouldn't be fiddled with outside parted, even though it's technically exposed. Could you please look at fixing parted to avoid clobbering the firmware area instead? I think that would be more correct. I have no idea what is really meant with the comment /* office use only ;-) */ int needs_clobber; in /usr/include/parted/disk.h, so I would like to ask upstream for clarification. Otavio, Jim: you are listed as parted upstream maintainers on http://www.gnu.org/software/parted/ - could you shed some light on this? Is it ok for an application to fiddle with the needs_clobber element or is this to be considered strictly libparted-internal? @Colin: If the solution is to be completely encapsulated in libparted, I see a large problem in how to solve the conflict between space after the partition table being very differently used by firmware for different SoCs on one side, and libparted trying to make sure that there are no remains of other partition table formats in the respective area on the other side - at least with the prerequisite of keeping both the current defaults (clobbering) as well as keeping the libparted API unchanged. With the current there is one erase size for all platforms method in ped_disk_clobber() in libparted/disk.c, we inevitably end up with conflicting requirements. An example: the source for ped_disk_clobber() states: /* How many sectors to zero out at each end. This must be large enough to zero out the magic bytes starting at offset 8KiB on a DASD partition table. Doing the same from the end of the disk is probably overkill, but at least on GPT, we do need to zero out the final sector. */ So for at least one platform (according to Wikipedia DASD seems to be some s/390 format), the area at offset 8KiB has to be wiped out while for another (armhf/sunxi) it may not be wiped out as the u-boot SPL is located there and cannot be relocated because its sector address is hardcoded in the SoC. According to https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=751704#25, similar problems (but at other offsets) come up for other SoCs. According to the examples listed there, for TI SoCs libparted would have to stop clobbering the GPT header after writing a DOS MBR, but could wipe the DASD area. For a Freescale i.MX SoC libparted could legally clobber the GPT header, but would have to refrain from clobbering the areas behind it. If you extrapolate this to the large number of different SoC families, to handle this completely inside libparted, the library would have to know what exactly is the target system for which it writes a partition table and special-case the handling for each of them. While one might assume that the partition table is for an s/390 system when a DASD partition table is generated, this does not work as easily for the plethora of different architectures and systems that use DOS MBR partition tables. This gets even more complicated by the fact that libparted may run on one platform but modify partition tables for another platform, such as when operating on disk images for use with emulators or when operating on hd-media images for different arm platforms (with different firmware/bootloader requirements) on one autobuild host, so trying to do runtime detection of the host system would still not cover all use cases. In the case of creating images from scratch, the problem can be circumvented by (re-)writing the bootloader at the end of the process, but when the task is to modify existing images with unknown content, this workaround is not available. As a conclusion, I think that the decision whether to clobber the area between the partition table and the beginning of the first partition has to be taken by the calling application and not internally by the library. Kind Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble?
Bug#751704: partman-base 173: partman overwrites parts of u-boot
Control: tag -1 patch On Wed, Jun 18, 2014 at 09:44:12PM +0200, Karsten Merker wrote: [On sunxi-based systems, upon writing the partition table, partman overwrites parts of u-boot which are located between the end of the partition table and the beginning of the first partition.] Hello, the following patch solves the issue for me. I have run a successful jessie installation on a Cubietruck with this patch applied to a locally built d-i based on kernel 3.16 from experimental. If there are no objections, I would like apply it to the partman-base git repository. Regards, Karsten From 35d41e3bfce60fd08c4da6dc0696a479c78bdcdd Mon Sep 17 00:00:00 2001 From: Karsten Merker mer...@debian.org Date: Tue, 12 Aug 2014 23:10:13 +0200 Subject: [PATCH] Take care of the firmware area on sunxi-based systems By default partman calls ped_disk_clobber when writing the partition table, but on sunxi-based systems this would overwrite the firmware area, resulting in an unbootable system (see bug #751704). Handle this as a special case in command_commit(). --- parted_server.c | 34 ++ 1 file changed, 34 insertions(+) diff --git a/parted_server.c b/parted_server.c index 55cf151..bc4523a 100644 --- a/parted_server.c +++ b/parted_server.c @@ -1330,6 +1330,25 @@ command_dump() oprintf(OK\n); } +/* Check whether we are running on a sunxi-based system. */ +int +is_sunxi_system() +{ +int cpuinfo_handle; +int result=0; +char buf[4096]; +int length; + +if ((cpuinfo_handle = open(/proc/cpuinfo, O_RDONLY)) != -1 ) { +length = read(cpuinfo_handle, buf, sizeof(buf)-1); +buf[length]='\0'; +if (strstr(buf,Allwinner)!=NULL) +result=1; +close(cpuinfo_handle); +} +return result; +} + void command_commit() { @@ -1337,6 +1356,21 @@ command_commit() if (dev == NULL) critical_error(The device %s is not opened., device_name); log(command_commit()); + +/* + * The boot device on sunxi-based systems needs special handling. + * By default partman calls ped_disk_clobber when writing the + * partition table, but on sunxi-based systems this would overwrite + * the firmware area, resulting in an unbootable system (see + * bug #751704). + */ +if (is_sunxi_system() !strcmp(disk-dev-path, /dev/mmcblk0)) { +disk-needs_clobber=0; +log(Sunxi platform detected. Disabling ped_disk_clobber \ +for the boot device %s to protect the firmware \ +area., disk-dev-path); +} + open_out(); if (disk != NULL named_is_changed(device_name)) ped_disk_commit(disk); -- 2.1.0.rc1 -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#751704: partman-base 173: partman overwrites parts of u-boot
Source: partman-base Version: 173 Severity: important Upon testing a locally built debian-installer based on linux 3.15-1 (from experimental) on an Allwinner sunXi-based armhf system, I have found that the system does not boot anymore after partman has written a new partition table to the SD card from which the system loads u-boot. Kernel 3.15-1 is the first kernel in Debian that contains an MMC/SD driver for this platform, i.e. with older versions partitions could only be created on an attached SATA disk but not on the SD card, so this problem has not shown up earlier. In my first attempt, the partition table has been created by the guided partitioning option in d-i, which has created the following layout: Disk /dev/sdd: 15 GB, 15718510080 bytes 255 heads, 63 sectors/track, 1911 cylinders, total 30700215 sectors Units = sectors of 1 * 512 = 512 bytes Device Boot Start End Blocks Id System /dev/sdd1 *2048 499711 257008 83 Linux Warning: Partition 1 does not end on cylinder boundary. /dev/sdd2 4997122930687914402272 83 Linux Warning: Partition 2 does not end on cylinder boundary. /dev/sdd32930892630701567 6988275 Extended Warning: Partition 3 does not end on cylinder boundary. /dev/sdd52930892830701567 698827 82 Linux swap Warning: Partition 5 does not end on cylinder boundary. The layout itself appears ok as the first partition starts at sector 2048, i.e. at 1MB offset, but comparing the state of the first MB of data on the device before and after partman has written the partition table shows that not only the partition table has been written, but also the area from offset 0x2000 up to 0x25ff has been zeroed out. On all Allwinner sunXi-based systems this area contains the u-boot SPL, without which the system cannot boot anymore. A second attempt with manual partitioning showed the same behaviour, i.e. it appears to be a general partman problem and not specific to partman-auto. Partman should only write the partition table but should not touch the area between the partition table and the beginning of the first partition as this area is used by bootloaders. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#751012: [flash-kernel] Flash-kernel should fail with exit 0 under debrootstrap
On Mon, Jun 09, 2014 at 05:54:35PM +0200, Bastien ROUCARIES wrote: On Mon, Jun 9, 2014 at 5:21 PM, Karsten Merker mer...@debian.org wrote: On Mon, Jun 09, 2014 at 03:48:11PM +, bastien ROUCARIES wrote: Package: flash-kernel Severity: important Flash-kernel should detect it run under debrootstrap and fail gracefully/ Install message: Processing triggers for libc-bin (2.18-7) ... Processing triggers for initramfs-tools (0.115) ... update-initramfs: Generating /boot/initrd.img-3.14-1-kirkwood /bin/df: Warning: cannot read table of mounted file systems: No such file or directory warning: failed to read mtab ^Cdpkg: error processing package initramfs-tools (--configure): subprocess installed post-installation script was interrupted Errors were encountered while processing: initramfs-tools E: Sub-process /usr/bin/dpkg returned an error code (1) Hello, I am trying to understand which problem exactly you have encountered, but I am a bit confused: you have filed a bug against flash-kernel but in the log you have provided, it is not flash-kernel that shows an error message, but initramfs-tools. I also cannot see in which context that has happened and how it relates to debootstrap - by default debootstrap does neither install a kernel image nor flash-kernel. From your log, it looks like you have manually interrupted the update-initramfs process, resulting in the error message above: No I have not interupted. I have installed te kirkwood kernel and flash-kernel on my chroot. Hello, I unfortunately cannot yet reproduce your problem. Due to the kernel version (3.14-1) I assume that you are debootstrapping jessie or sid. I have just debootstrapped a sid/armel chroot that included flash-kernel and linux-image-3.14-1-kirkwood on a sid/armhf system without problems: # debootstrap --arch=armel --include=flash-kernel,linux-image-kirkwood sid armel-sid-chroot http://ftp.de.debian.org/debian I: Retrieving Release I: Retrieving Release.gpg I: Checking Release signature I: Valid Release signature (key id A1BD8E9D78F7FE5C3E65D8AF8B48AD6246925553) [...] I: Configuring linux-image-3.14-1-kirkwood... I: Configuring flash-kernel... I: Configuring libgnutls-openssl27:armel... I: Configuring wget... I: Configuring libcwidget3:armel... I: Configuring aptitude... I: Configuring linux-image-kirkwood... I: Configuring iputils-ping... I: Configuring tasksel... I: Configuring tasksel-data... I: Configuring perl-modules... I: Configuring perl... I: Configuring init-system-helpers... I: Configuring cron... I: Configuring rsyslog... I: Configuring logrotate... I: Configuring libc-bin... I: Configuring initramfs-tools... I: Base system installed successfully. # Just to be sure: On which kind of hardware (kirkwood or non-kirkwood system) and in which Debian release (wheezy, jessie or sid) are you running the debootstrap command and which release are you bootstrapping with debootstrap (jessie or sid)? Have you created your chroot like I did above, i.e. with the --include parameter, or have you first run debootstrap without it and later on manually installed linux-image-3.14-1-kirkwood and flash-kernel in the already-created chroot? Please provide a full log of the whole process starting with the invocation of debootstrap up to the point where flash-kernel runs but should not. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#745972: Please enable additional kernel config options for sunxi support (starting with kernel 3.14)
Package: linux Severity: wishlist Hello, I am working on getting ARM platforms based on the Allwinner A10 (sun4i) and A20 (sun7i) SOCs better supported in Debian. During the last months quite a bit of drivers for these SOCs have been integrated into the mainline Linux kernel. I would therefore like to ask for the following additional kernel configuration options to be enabled in future Debian kernel packages: Kernel 3.14 === CONFIG_RTC_DRV_SUNXI=m (sunxi realtime clock support) CONFIG_USB_EHCI_HCD_PLATFORM=m (platform device for enabling the embedded EHCI controller) CONFIG_USB_OHCI_HCD_PLATFORM=m (platform device for enabling the embedded OHCI controller) CONFIG_SUNXI_WATCHDOG=m(embedded watchdog device) Kernel 3.15 (drivers accepted in 3.15rc1) === CONFIG_PHY_SUN4I_USB=m (PHY driver for the embedded OHCI and EHCI controllers) CONFIG_SPI=y CONFIG_SPI_SUNXI=m (sunxi SPI master driver) Kernel 3.16 (driver submitted and currently in review) === CONFIG_MMC_SUNXI=m (sunxi MMC/SD controller driver) Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#744365: lb config does not create config directory
Package: live-build Version: 4.0~alpha33-1 Tags: patch Hello, running lb config from the current unstable live-build package (4.0~alpha33-1, identical to current git as of commit 145418141e4965691dad1757c6fbdffa4a96290c) in an empty directory results in P: Creating config tree for a debian/wheezy/amd64 system /usr/lib/live/build/config: 892: /usr/lib/live/build/config: cannot create config/common: Directory nonexistent The code fragment responsible for this is: 884 if [ ! -e config ] 885 then 886 Echo_message Creating config tree for a ${LB_MODE}/${LB_DISTRIBUTION}/${LIVE_IMAGE_ARCHITECTURE} system 887 else 888 Echo_message Updating config tree for a ${LB_MODE}/${LB_DISTRIBUTION}/${LIVE_IMAGE_ARCHITECTURE} system 889 fi 890 891 # Creating live-build configuration 892 cat config/common EOF The if clause in line 884 checks for the existence of the config directory, but does not create it when missing, so trying to create config/common in line 892 fails when running lb config without an existing config directory. The following one-line patch fixes that. diff --git a/scripts/build/config b/scripts/build/config index 5261c65..9430c5e 100755 --- a/scripts/build/config +++ b/scripts/build/config @@ -884,6 +884,7 @@ Check_defaults if [ ! -e config ] then Echo_message Creating config tree for a ${LB_MODE}/${LB_DISTRIBUTION}/${LIVE_IMAGE_ARCHITECTURE} system + mkdir -p config else Echo_message Updating config tree for a ${LB_MODE}/${LB_DISTRIBUTION}/${LIVE_IMAGE_ARCHITECTURE} system fi Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#735092: flash-kernel: add Raspberry Pi support
Package: flash-kernel Version: 3.11 Severity: wishlist Tags: patch The attached patch adds Raspberry Pi support to flash-kernel. It is based on current flash-kernel git (as of 7f52719ab0a607b89555baffd1cc8c14207c0f8f). It is available for merging in the rpi-support-rebased branch at http://anonscm.debian.org/gitweb/?p=users/merker/flash-kernel.git;a=shortlog;h=refs/heads/rpi-support-rebased Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. From ff2103a10fc27f77c2763d9dd4b043254b6e489d Mon Sep 17 00:00:00 2001 From: K. Merker mer...@debian.org Date: Wed, 1 Jan 2014 13:17:13 +0100 Subject: [PATCH] Initial Raspberry Pi support Add a method rpiconfigtxt and a new machine database entry for the Raspberry Pi. The method rpiconfigtxt enables flash-kernel to update the kernel file name to be booted by the Raspberry Pi firmware. As the Raspberry Pi firmware can only boot from a FAT partition, using symlinks to point to the kernel to be booted is not possible. Therefore the kernel filename entry in the firmware config has to be changed after installation of a new kernel package. --- README| 10 +- db/all.db |5 + functions | 19 +++ test_db |2 +- 4 files changed, 34 insertions(+), 2 deletions(-) diff --git a/README b/README index 51aa15f..c7cde61 100644 --- a/README +++ b/README @@ -49,6 +49,7 @@ The following systems are supported: - QNAP TS-409 - QNAP TS-410 and TS-410U Turbo NAS - QNAP TS-419P and TS-419U Turbo NAS + - Raspberry Pi - Seagate FreeAgent DockStar - SheevaPlug - SheevaPlug eSATA @@ -153,12 +154,19 @@ The supported fields are: * Method: (optional) indicates how to support this particular machine; the default is generic; other available methods are: android, multi, - redboot, slug, symlink + redboot, slug, symlink, rpiconfigtxt * Boot-Device: (optional) block device to mount before installing kernel, initrd and U-Boot script; Boot-Kernel-Path, Boot-Initrd-Path and Boot-Script-Path are then taken relative to this boot device +* Rpi-ConfigTxt-Path: (optional) Raspberry Pi firmware configuration pathname. + The Raspberry Pi firmware configuration is stored in a text file on the + first (FAT formatted) partition of the SD card in the system and + contains settings like kernel file name, video mode, memory split + between CPU and GPU, overclocking parameters, etc. Rpi-ConfigTxt-Path + contains the full path to this file (default value: /boot/config.txt). + Configuration - - - - - - - diff --git a/db/all.db b/db/all.db index c28432a..c0fdf27 100644 --- a/db/all.db +++ b/db/all.db @@ -382,3 +382,8 @@ Method: android Android-Boot-Device: /dev/mmcblk0 Required-Packages: abootimg Bootloader-Sets-Root: no + +Machine: BCM2708 +Method: rpiconfigtxt +Rpi-ConfigTxt-Path: /boot/config.txt +Bootloader-Sets-Root: yes diff --git a/functions b/functions index 1233981..26a16ed 100644 --- a/functions +++ b/functions @@ -440,6 +440,7 @@ boot_script_path=$(get_machine_field $machine Boot-Script-Path) || : boot_dtb_path=$(get_machine_field $machine Boot-DTB-Path) || : boot_multi_path=$(get_machine_field $machine Boot-Multi-Path) || : android_boot_device=$(get_machine_field $machine Android-Boot-Device) || : +rpi_configtxt_path=$(get_machine_field $machine Rpi-ConfigTxt-Path) || : if [ -n $dtb_append_from ]; then if dtb_append_required; then @@ -711,6 +712,24 @@ case $method in } $imtd || error failed. echo done. 2 ;; + rpiconfigtxt) + rpi_configtxt_path=${rpi_configtxt_path:-/boot/config.txt} + rpi_configtxt_tempfile=$(tempfile) + + if [ ! -e ${rpi_configtxt_path} ]; then + echo ${rpi_configtxt_path} not found. Creating it. + touch ${rpi_configtxt_path} + fi + + grep -E -v ^#fk-old- ${rpi_configtxt_path} | \ + sed -e 's/^kernel=/#fk-old-kernel=/' \ + -e 's/^initramfs/#fk-old-initramfs/' \ + -e 's/^ramfsfile/#fk-old-ramfsfile/' \ + -e 's/^ramfsaddr/#fk-old-ramfsaddr/' ${rpi_configtxt_tempfile} + + echo -n kernel=$(basename ${kfile})\ninitramfs $(basename ${ifile})\n ${rpi_configtxt_tempfile} + mv ${rpi_configtxt_tempfile} ${rpi_configtxt_path} + ;; esac } diff --git a/test_db b/test_db index aec83f1..217fe64 100755 --- a/test_db +++ b/test_db @@ -22,7 +22,7 @@ MACHINE_DB=$(cat ${FK_CHECKOUT:-$FK_DIR}/db/*.db) test_no_unknown_fields() { -local expected='Android-Boot-Device Boot-Device Boot-DTB-Path Boot-Initrd-Path Boot-Kernel-Path Boot-Multi-Path Boot-Script-Path Bootloader-Sets-Root DTB-Append DTB-Append-From DTB-Id Kernel-Flavors Machine Machine-Id Method Mtd-Initrd Mtd-Kernel Optional-Packages Required-Packages U-Boot-Initrd-Address U-Boot-Kernel-Address U-Boot-Kernel-Entry-Point U-Boot-Multi-Address U-Boot-Script-Address U-Boot-Script-Name' +local expected='Android-Boot-Device Boot-Device
Bug#735093: flash-kernel: does not correctly handle removal of the highest-versioned installed kernel package
Package: flash-kernel Version: 3.11 Severity: normal Tags: patch Flash-Kernel 3.11 does not properly handle a removal of the highest-versioned kernel package. The same applies to current git as of 2014-01-12 which will probably be released as 3.12. When called from the kernel postinst in this case, it just aborts instead of flashing the remaining then-highest-versioned kernel, which in specific cases could lead to an unbootable system. Attached is a patch against current mainline flash-kernel git (as of 7f52719ab0a607b89555baffd1cc8c14207c0f8f). It is available for merging in the rpi-support-rebased branch at http://anonscm.debian.org/gitweb/?p=users/merker/flash-kernel.git;a=shortlog;h=refs/heads/rpi-support-rebased Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. From 989312441af9bdec76ba571eb3c6ae3d65f93d4e Mon Sep 17 00:00:00 2001 From: K. Merker mer...@debian.org Date: Thu, 2 Jan 2014 14:34:32 +0100 Subject: [PATCH] Add extended kernel package removal handling Until now, flash-kernel has not properly handled a removal of the highest-versioned kernel package. When called from the kernel postinst in this case, it just aborted instead of flashing the remaining then-highest-versioned kernel, which could lead to an unbootable system. This patch adds extended removal handling which takes care of the issue. --- debian/copyright|1 + flash-kernel.8 | 18 ++--- functions | 63 +++ kernel-hook/zz-flash-kernel |6 +++-- 4 files changed, 71 insertions(+), 17 deletions(-) diff --git a/debian/copyright b/debian/copyright index af21da0..e9fb23c 100644 --- a/debian/copyright +++ b/debian/copyright @@ -6,6 +6,7 @@ Copyright: Copyright (C) 2006 Joey Hess jo...@debian.org Copyright (C) 2006, 2007, 2008, 2009, 2010, 2011 Martin Michlmayr t...@cyrius.com Copyright (C) 2011 Loïc Minier l...@dooz.org +Copyright (C) 2014 Karsten Merker mer...@debian.org This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License diff --git a/flash-kernel.8 b/flash-kernel.8 index 7b5bad5..8165372 100644 --- a/flash-kernel.8 +++ b/flash-kernel.8 @@ -3,7 +3,7 @@ .SH NAME flash-kernel \- put kernel and initramfs in boot location .SH SYNOPSIS -.B flash-kernel [--supported] [kvers] +.B flash-kernel [--machine machine-type] [--context calling-context] [--supported] [kvers] .SH DESCRIPTION flash-kernel is a script which will put the kernel and initramfs in the boot location of embedded devices that don't load the kernel and @@ -11,7 +11,7 @@ initramfs directly from /boot. flash-kernel supports devices that boot from flash memory (hence the name) as well as some devices that require a special boot image on the disk. .P -Optionally, it can be passed a version of the kernel to flash; only +Optionally, it can be passed a version of the kernel to process; only the highest version will be flashed and other versions will be ignored. Kernel and initrd are read from /boot. .P @@ -20,7 +20,17 @@ the subarchitectures of the machine and the image to be flashed match. Valid filenames for images to flash are suffixed with the subarchitecture. .P -If the \-\-supported option is used, flash\-kernel will test to see if +If the \-\-supported option is used, flash\-kernel will test whether the hardware is supported, and return a true or false value. +.P +The \-\-machine option allows to manually override the auto-detected +machine type string. +.P +The \-\-context option is used internally by the kernel +postinst/postrm scripts which call flash-kernel upon kernel package +installations and removals. It provides flash-kernel with the +information that it is running in the kernel packages +postinst/postrm context. Valid values are postinst.d* and +postrm.d*. .SH AUTHOR -Martin Michlmayr t...@cyrius.com +Martin Michlmayr t...@cyrius.com, Karsten Merker mer...@debian.org \ No newline at end of file diff --git a/functions b/functions index 26a16ed..4b1745e 100644 --- a/functions +++ b/functions @@ -351,28 +351,69 @@ android_flash() { } main() { -if [ x$1 = x--machine ]; then - machine=$2 - shift 2 -else + +while [ $# -gt 0 ] +do +if [ x$1 = x--machine ]; then +machine=$2 +shift 1 +elif [ x$1 = x--context ]; then +context=$2 +shift 1 +elif [ x$1 = x--supported ]; then +do_check_supported=true +elif [ $# -eq 1 ]; then +kvers=$1 +fi +shift 1 +done + +if [ -z $machine ]; then get_machine fi -if [ x$1 = x--supported ]; then +if [ -n $do_check_supported ]; then if check_supported $machine; then exit 0 + else + exit 1 fi - exit 1 fi -# kernel
Bug#735092: flash-kernel: add Raspberry Pi support
On Sun, Jan 12, 2014 at 07:01:36PM +, Ian Campbell wrote: On Sun, 2014-01-12 at 19:24 +0100, Karsten Merker wrote: On Sun, Jan 12, 2014 at 06:00:24PM +, Ben Hutchings wrote: On Sun, 2014-01-12 at 18:34 +0100, Karsten Merker wrote: The attached patch adds Raspberry Pi support to flash-kernel. [SNIP] So the boot partition is mounted at /boot? Doesn't that make it impossible to install packaged kernels (as dpkg needs to create hard links)? Installing a kernel package (locally built using make-kpkg) with a VFAT /boot does not show any problems here: The normal way of dealing with this is to not have the bootloader partition mounted on boot, this is described in flash-kernel via the Boot-Device field. I think this should be honoured for Rpi too. Rpi-ConfigTxt-Path would then be relative to this device. Having the vfat boot partition as /boot is the standard way in Raspbian (Debian/armhf rebuild for the Pi/armv6, official OS distribution for the Pi) and is described this way in all user documentation for the Pi. This is also expected by other software, in particular the firmware update mechanism, which expects the firmware image to be at /boot/bootcode.bin, as well as by the Raspberry Pi foundation's kernel updates (which do not come as a Debian kernel package). Not mounting the vfat boot partition at /boot would of course be technically possible, but nobody actually does that on a Pi and it would pose several problems: - The firmware update mechanism would have to be changed. This is outside the scope of Debian and as installing firmware updates is something that happens rather frequently on the Pi, that must work without problems. - All the Pi user documentation would have to be changed, which is also outside the scope of Debian. - In case the vfat boot partition is not mounted at /boot, the firmware cannot anymore directly boot the /boot/vmlinuz-* and /boot/initrd.img-* files from the kernel package. That would therefore require copying the files from /boot to the actual vfat boot partition. As having the vfat boot partition mounted on /boot is the actual default case, flash-kernel would have to handle this as well. This means either special handling depending on whether the vfat partition is at /boot or not, or generally copying the kernel to the vfat partition under another name. The latter would in the case vfat-at-/boot result in having the kernel and initrd twice on the already quite small system vfat partition, which is a route I would like to avoid. All things taken together I think that we should simply accept that the vfat boot partition gets mounted at /boot on the Pi and use it this way. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#735092: flash-kernel: add Raspberry Pi support
On Sun, Jan 12, 2014 at 07:32:56PM +, Ben Hutchings wrote: On Sun, 2014-01-12 at 19:24 +0100, Karsten Merker wrote: On Sun, Jan 12, 2014 at 06:00:24PM +, Ben Hutchings wrote: So the boot partition is mounted at /boot? Doesn't that make it impossible to install packaged kernels (as dpkg needs to create hard links)? Installing a kernel package (locally built using make-kpkg) with a VFAT /boot does not show any problems here: [...] Please also test upgrading (or simply reinstalling the same package again). Indeed, reinstalling fails: dpkg: error processing linux-image-3.12.6-rpi+_2014010201_armhf.deb (--install): unable to make backup link of `./boot/vmlinuz-3.12.6-rpi+' before installing new version: Operation not permitted Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#728014: [PATCH] Fix cross-building armhf kernel packages
Package: kernel-package Version: 12.036+nmu3 Tags: patch I am rather new to the arm architecture and currently trying to get a couple of things working on a Raspberry Pi. Building kernel packages natively on a 700MHz single-core ARMv6 cpu takes ages, so I have been looking into possibilities for cross-building kernel packages from faster architectures. According to the make-kpkg manpage, the following command $ make-kpkg --arch armhf --cross_compile arm-rpi-linux-gnueabi- \ --revision 20131002 --append_to_version -rpi --initrd \ --us --uc --rootcmd fakeroot kernel_image should build an armhf kernel package on an amd64 host with an arm-rpi-linux-gnueabi-crosstoolchain installed. With kernel-package 12.036+nmu3 this fails with debian/ruleset/misc/checks.mk:36: *** Error. I do not know where the kernel image goes to [kimagedest undefined] The usual case for this is that I could not determine which arch or subarch this machine belongs to. Please specify a subarch, and try again.. Stop. It looks like there are just a few definitions for armhf missing, which are only used in the case of cross-building - native builds work fine. With the attached patch, cross-building kernel packages for armhf works for me, i.e. it results in working kernel-image packages for the Raspberry Pi. I had asked on debian-arm whether anybody sees issues with the patch, but have received no reply. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. diff -Nur kernel-package-12.036+nmu3/kernel/ruleset/arches/armhf.mk kernel-package-12.036+nmu3.patched/kernel/ruleset/arches/armhf.mk --- kernel-package-12.036+nmu3/kernel/ruleset/arches/armhf.mk 1970-01-01 01:00:00.0 +0100 +++ kernel-package-12.036+nmu3.patched/kernel/ruleset/arches/armhf.mk 2013-10-02 22:02:47.0 +0200 @@ -0,0 +1,34 @@ +# -*- Mode: Makefile-Gmake -*- +## armhf.mk --- +## +## This program is free software; you can redistribute it and/or modify +## it under the terms of the GNU General Public License as published by +## the Free Software Foundation; either version 2 of the License, or +## (at your option) any later version. +## +## This program is distributed in the hope that it will be useful, +## but WITHOUT ANY WARRANTY; without even the implied warranty of +## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +## GNU General Public License for more details. +## +## You should have received a copy of the GNU General Public License +## along with this program; if not, write to the Free Software +## Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA +## +### + +### ARMHF +ifeq ($(strip $(architecture)),armhf) + kimage := vmlinuz + target = zImage + NEED_DIRECT_GZIP_IMAGE=NO + kimagesrc = arch/$(KERNEL_ARCH)/boot/zImage + kimagedest = $(INT_IMAGE_DESTDIR)/vmlinuz-$(KERNELRELEASE) + DEBCONFIG = $(CONFDIR)/config.armhf + kelfimagesrc = vmlinux + kelfimagedest = $(INT_IMAGE_DESTDIR)/vmlinux-$(KERNELRELEASE) +endif + +#Local variables: +#mode: makefile +#End: diff -Nur kernel-package-12.036+nmu3/kernel/ruleset/architecture.mk kernel-package-12.036+nmu3.patched/kernel/ruleset/architecture.mk --- kernel-package-12.036+nmu3/kernel/ruleset/architecture.mk 2012-03-04 02:22:23.0 +0100 +++ kernel-package-12.036+nmu3.patched/kernel/ruleset/architecture.mk 2013-10-02 22:01:08.0 +0200 @@ -59,6 +59,10 @@ ifeq ($(strip $(architecture)),armeb) include $(DEBDIR)/ruleset/arches/armeb.mk endif +ifeq ($(strip $(architecture)),armhf) +include $(DEBDIR)/ruleset/arches/armhf.mk +endif + # PowerPC and PowerPC architecture ifneq ($(strip $(filter ppc powerpc ppc64 powerpc64,$(architecture))),) diff -Nur kernel-package-12.036+nmu3/kernel/ruleset/misc/kernel_arch.mk kernel-package-12.036+nmu3.patched/kernel/ruleset/misc/kernel_arch.mk --- kernel-package-12.036+nmu3/kernel/ruleset/misc/kernel_arch.mk 2012-03-04 02:39:27.0 +0100 +++ kernel-package-12.036+nmu3.patched/kernel/ruleset/misc/kernel_arch.mk 2013-10-02 21:55:54.0 +0200 @@ -56,6 +56,10 @@ KERNEL_ARCH := arm endif +ifeq ($(strip $(architecture)),armhf) + KERNEL_ARCH := arm +endif + ifeq ($(strip $(architecture)),hppa) KERNEL_ARCH := parisc endif
Bug#706583: Installation report Wheezy DI rc2 amd64 xfce: installation ok
Package: installation-reports Boot method: isohybrid image booted from USB stick Image version: http://cdimage.debian.org/cdimage/wheezy_di_rc2/amd64/iso-cd/debian-wheezy-DI-rc2-amd64-netinst.iso Date: 2013-05-01 Mainboard: ECS P35T-A Processor: Intel E3300 Memory: 4GB Output of lspci -knn (or lspci -nn): 00:00.0 Host bridge [0600]: Intel Corporation 82G33/G31/P35/P31 Express DRAM Controller [8086:29c0] (rev 02) 00:01.0 PCI bridge [0604]: Intel Corporation 82G33/G31/P35/P31 Express PCI Express Root Port [8086:29c1] (rev 02) 00:19.0 Ethernet controller [0200]: Intel Corporation 82566DC-2 Gigabit Network Connection [8086:294c] (rev 02) 00:1a.0 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #4 [8086:2937] (rev 02) 00:1a.1 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #5 [8086:2938] (rev 02) 00:1a.7 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #2 [8086:293c] (rev 02) 00:1b.0 Audio device [0403]: Intel Corporation 82801I (ICH9 Family) HD Audio Controller [8086:293e] (rev 02) 00:1c.0 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 1 [8086:2940] (rev 02) 00:1c.1 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 2 [8086:2942] (rev 02) 00:1c.2 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 3 [8086:2944] (rev 02) 00:1c.3 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 4 [8086:2946] (rev 02) 00:1c.4 PCI bridge [0604]: Intel Corporation 82801I (ICH9 Family) PCI Express Port 5 [8086:2948] (rev 02) 00:1d.0 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #1 [8086:2934] (rev 02) 00:1d.1 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #2 [8086:2935] (rev 02) 00:1d.2 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #3 [8086:2936] (rev 02) 00:1d.3 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB UHCI Controller #6 [8086:2939] (rev 02) 00:1d.7 USB controller [0c03]: Intel Corporation 82801I (ICH9 Family) USB2 EHCI Controller #1 [8086:293a] (rev 02) 00:1e.0 PCI bridge [0604]: Intel Corporation 82801 PCI Bridge [8086:244e] (rev 92) 00:1f.0 ISA bridge [0601]: Intel Corporation 82801IH (ICH9DH) LPC Interface Controller [8086:2912] (rev 02) 00:1f.2 IDE interface [0101]: Intel Corporation 82801IR/IO/IH (ICH9R/DO/DH) 4 port SATA Controller [IDE mode] [8086:2920] (rev 02) 00:1f.3 SMBus [0c05]: Intel Corporation 82801I (ICH9 Family) SMBus Controller [8086:2930] (rev 02) 00:1f.5 IDE interface [0101]: Intel Corporation 82801I (ICH9 Family) 2 port SATA Controller [IDE mode] [8086:2926] (rev 02) 01:00.0 VGA compatible controller [0300]: Advanced Micro Devices [AMD] nee ATI RV505 [Radeon X1550 Series] [1002:7143] 01:00.1 Display controller [0380]: Advanced Micro Devices [AMD] nee ATI RV505 [Radeon X1550 Series] (Secondary) [1002:7163] 06:00.0 SATA controller [0106]: JMicron Technology Corp. JMB361 AHCI/IDE [197b:2361] (rev 02) 06:00.1 IDE interface [0101]: JMicron Technology Corp. JMB361 AHCI/IDE [197b:2361] (rev 02) Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[O] Configure network: [O] Detect CD: [O] Load installer modules: [O] Detect hard drives: [O] Partition hard drives: [O] Install base system:[O] Clock/timezone setup: [O] User/password setup:[O] Install tasks: [O] Install boot loader:[O] Overall install:[O] Comments/Problems: The installation process worked flawlessly. Installation was done using a network mirror; network connectivity was provided via a TP-Link TL-WN727N 802.11n WLAN dongle (USB ID: 148f:5370, Ralink RT5370 chipset). The installer prompted for the WLAN dongle's firmware, which was provided on a USB stick (downloaded from http://cdimage.debian.org/cdimage/unofficial/non-free/firmware/wheezy/current/). Basic X11 functionality worked out of the box, advanced X11 features (in particular XVideo support) required installation of the non-free radeon firmware from the package firmware-linux-nonfree. A sidenote: while the WLAN via the TL-WN727N generally works, interactive use (in particular using logins via ssh) shows that the connection often has short lags or hangs. This happens even when the WLAN dongle is within 2m of the access point with direct line of sight. Regards, Karsten -- Gem. Par. 28 Abs. 4 Bundesdatenschutzgesetz widerspreche ich der Nutzung sowie der Weitergabe meiner personenbezogenen Daten für Zwecke der Werbung sowie der Markt- oder Meinungsforschung. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#683692: installation report: [i386] [daily 2012/08/02] installation hangs forever
Package: installation-reports Boot method: credit card image booted from USB key Image version: http://cdimage.debian.org/cdimage/daily-builds/daily/arch-latest/i386/jigdo-cd/debian-testing-i386-netinst.jigdo Image date: 2012/08/02 17:27 Installation date: 2012/08/02 Machine: Desktop PC Processor: Pentium DualCore E2200 Memory: 2 GB Partitions: installer fails before partitioning stage Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[E] Configure network: [ ] Detect CD: [ ] Load installer modules: [ ] Detect hard drives: [ ] Partition hard drives: [ ] Install base system:[ ] Clock/timezone setup: [ ] User/password setup:[ ] Install tasks: [ ] Install boot loader:[ ] Overall install:[ ] Comments/Problems: The machine has a TP-Link TL-WN721N 802.11n WLAN dongle (based on an Atheros AR9271 chipset) connected via USB. As long as this device is connected to the computer, the detect network stage just hangs forever. If the WLAN dongle is not connected to the machine, d-i works as expected. The firmware for the WLAN dongle is non-free and therefore not included in d-i, but the installation should not hang completely just because the device is connected to the computer. The device itself is supported by Linux and works when connected after the installation is finished, provided the (non-free) package firmware-atheros has been installed. Relevant syslog entries: Aug 2 19:03:39 kernel: [2.448030] usb 1-3: new high-speed USB device number 4 using ehci_hcd Aug 2 19:03:39 kernel: [2.604889] usb 1-3: New USB device found, idVendor=0cf3, idProduct=9271 Aug 2 19:03:39 kernel: [2.611200] usb 1-3: New USB device strings: Mfr=16, Product=32, SerialNumber=48 Aug 2 19:03:39 kernel: [2.617486] usb 1-3: Product: USB2.0 WLAN Aug 2 19:03:39 kernel: [2.623615] usb 1-3: Manufacturer: ATHEROS Aug 2 19:03:39 kernel: [2.629689] usb 1-3: SerialNumber: 12345 Aug 2 19:03:49 hw-detect: Missing modules 'eth1394 (FireWire ethernet) Aug 2 19:03:50 check-missing-firmware: /dev/.udev/firmware-missing does not exist, skipping Aug 2 19:03:50 check-missing-firmware: /run/udev/firmware-missing does not exist, skipping Aug 2 19:03:50 check-missing-firmware: no missing firmware in /dev/.udev/firmware-missing /run/udev/firmware-missing Aug 2 19:04:59 kernel: [ 85.337797] cfg80211: Calling CRDA to update world regulatory domain Aug 2 19:04:59 net/hw-detect.hotplug: Detected hotpluggable network interface lo Aug 2 19:04:59 kernel: [ 85.424081] usb 1-3: ath9k_htc: Firmware - htc_9271.fw not found Aug 2 19:04:59 kernel: [ 85.424092] ath9k_htc: probe of 1-3:1.0 failed with error -22 Aug 2 19:04:59 kernel: [ 85.424112] usbcore: registered new interface driver ath9k_htc Aug 2 19:04:59 net/hw-detect.hotplug: Detected hotpluggable network interface eth0 Aug 2 19:04:59 hw-detect: Missing modules 'eth1394 (FireWire ethernet) Aug 2 19:05:00 check-missing-firmware: /dev/.udev/firmware-missing does not exist, skipping Aug 2 19:05:00 check-missing-firmware: missing firmware files (htc_9271.fw) for ath9k_htc Aug 2 19:05:01 kernel: [ 87.334257] FAT-fs (sda): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.363899] FAT-fs (sda): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.11] FAT-fs (sdc): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.519037] FAT-fs (sdc): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.584031] end_request: I/O error, dev fd0, sector 0 Aug 2 19:05:01 kernel: [ 87.608030] end_request: I/O error, dev fd0, sector 0 Aug 2 19:05:01 kernel: [ 87.632033] end_request: I/O error, dev fd0, sector 0 Aug 2 19:05:01 kernel: [ 87.656029] end_request: I/O error, dev fd0, sector 0 Aug 2 19:05:01 kernel: [ 87.675372] FAT-fs (sda1): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.705927] FAT-fs (sda1): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.833271] FAT-fs (sda): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.846341] FAT-fs (sda): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 87.926912] FAT-fs (sdc): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Aug 2 19:05:01 kernel: [ 88.001788] FAT-fs (sdc): utf8 is not a recommended IO charset for FAT