Planning/4.0

Release Schedule

Tentatively proposed dates, still to be discussed/agreed.

2018-12-12	Beginning of development phase
2019-03-12	Soft feature freeze. Only bug fixes after this point. All feature changes must be already in a sub maintainer tree and all pull requests from submaintainers must have been sent to the list by this date.
2019-03-19	Hard feature freeze. Tag rc0
2019-03-26	Tag rc1
2019-04-02	Tag rc2
2019-04-09	Tag rc3
2019-04-16	Release, or tag rc4 if extra RC cycle needed
2019-04-23	Release, if we needed an rc4.

Known issues

Please list all patch series or bugs which need to be fixed for the release here. That way I can be sure I don't miss any when I'm rolling release candidates.

Fixed in rc2

• spapr linux guests crash under TCG https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg06447.html
• TCG crash when running a ppc64le guest https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg07374.html
• aarch64-softmmu/virt unintentionally gained the "virtio-vga" device (via the Kconfig conversion), which misleads libvirt https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg06060.html -- fixed in commit 938912a86611

Fixed in rc3

• Pending migration PR https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg04457.html
• filemon inotify watch uniqneness https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg04885.html
• Travis timeout fixes https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg05775.html
• migration of intel_iommu due to root_scalable field
• MIPS FPU NaN propagation https://lists.gnu.org/archive/html/qemu-devel/2019-03/msg05808.html
• COLO bitmap_mutex error https://lists.nongnu.org/archive/html/qemu-devel/2019-03/msg07909.html
• migration to 3.1 gives : Unknown savevm section or instance 'audio' 0.
• linux-user compile is broken with latest glibc master, due to gettid clash
• PCIe extended config space not accessible on spapr https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg00204.html
• target/i386: Generate #UD for LOCK on a register increment
• Crash with qemu-system-x86_64 -monitor stdio -M pc-1.1 and device_add nvdimm,id=n1 - patch: https://patchwork.ozlabs.org/patch/1079964/
• roms: EFIROM variable clashes with iPXE https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg01033.html
• include/qemu/bswap.h: Use __builtin_memcpy() in accessor functions
• use-after-free in multifd_recv_unfill_packet() (CID 1400442)
• arithmetic overflow in load_device_tree()

Will be fixed in rc4

• broken naming of root PCI buses for spapr PHBs https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg019
• warning compiling curses.c on some 32-bit systems https://patchew.org/QEMU/20190402073018.17747-1-kraxel%40redhat.com/56.html
• qemu-img doc consistency https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg01548.html
• iotest 241 failure https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg01725.html
• iotest 245 failure https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg01682.html
• gcc -O3 buildfix in slirp code
• "[PATCH] socket: allow wait=false for client socket" (regression affects libvirt use of chardev) https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg02548.html

Not yet fixed in any rc

• "0/3 usb-mtp: fix ObjectInfo request handling" (potential guest-triggered security issue, read off end of buffer, only present if you're using usb-mtp and have explicitly enabled write support) https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg02524.html -- we might want to just apply patch 2 which is the security fix and looks safe to me

Issues that will not be fixed

• fix delvm interaction with libvirt https://lists.gnu.org/archive/html/qemu-devel/2019-04/msg01699.html (not a regression)
• allow seccomp to be used on archs other than x86/mips/ppc/arm/s390x (not a regression)

Targeted Features

See the ChangeLog/4.0 for full details.