| Message ID | 20220408103127.19219-1-varad.gautam@suse.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | SMP Support for x86 UEFI Tests | expand |
On Fri, Apr 08, 2022, Varad Gautam wrote: > This series brings multi-vcpu support to UEFI tests on x86. > > Most of the necessary AP bringup code already exists within kvm-unit-tests' > cstart64.S, and has now been either rewritten in C or moved to a common location > to be shared between EFI and non-EFI test builds. > > A call gate is used to transition from 16-bit to 32-bit mode, since EFI may > not load the 32-bit entrypoint low enough to be reachable from the SIPI vector. > > Git branch: https://github.com/varadgautam/kvm-unit-tests/commits/ap-boot-v1 > > Varad Gautam (9): > x86: Move ap_init() to smp.c > x86: Move load_idt() to desc.c > x86: desc: Split IDT entry setup into a generic helper > x86: efi, smp: Transition APs from 16-bit to 32-bit mode > x86: Move 32-bit bringup routines to start32.S > x86: efi, smp: Transition APs from 32-bit to 64-bit mode > x86: Move load_gdt_tss() to desc.c > x86: Provide a common 64-bit AP entrypoint for EFI and non-EFI > x86: setup: Serialize ap_start64 with a spinlock This series doesn't apply cleanly on upstream master. I feel bad for asking, but in addition to rebasing to master, can you also rebase on top of my series[*] that fixes SMP bugs that were introduced by the initial UEFI support? I don't think there will be semantic conflicts, but the whitespace cleanups (spaces => tabs) do conflict, and I'd really like to start purging the spaces mess from KUT. Paolo / Andrew, ping on my series, it still applies cleanly. [*] https://lore.kernel.org/all/20220121231852.1439917-1-seanjc@google.com
Hi Sean, On 4/8/22 4:35 PM, Sean Christopherson wrote: > On Fri, Apr 08, 2022, Varad Gautam wrote: >> This series brings multi-vcpu support to UEFI tests on x86. >> >> Most of the necessary AP bringup code already exists within kvm-unit-tests' >> cstart64.S, and has now been either rewritten in C or moved to a common location >> to be shared between EFI and non-EFI test builds. >> >> A call gate is used to transition from 16-bit to 32-bit mode, since EFI may >> not load the 32-bit entrypoint low enough to be reachable from the SIPI vector. >> >> Git branch: https://github.com/varadgautam/kvm-unit-tests/commits/ap-boot-v1 >> >> Varad Gautam (9): >> x86: Move ap_init() to smp.c >> x86: Move load_idt() to desc.c >> x86: desc: Split IDT entry setup into a generic helper >> x86: efi, smp: Transition APs from 16-bit to 32-bit mode >> x86: Move 32-bit bringup routines to start32.S >> x86: efi, smp: Transition APs from 32-bit to 64-bit mode >> x86: Move load_gdt_tss() to desc.c >> x86: Provide a common 64-bit AP entrypoint for EFI and non-EFI >> x86: setup: Serialize ap_start64 with a spinlock > > This series doesn't apply cleanly on upstream master. I feel bad for asking, but > in addition to rebasing to master, can you also rebase on top of my series[*] that > fixes SMP bugs that were introduced by the initial UEFI support? I don't think > there will be semantic conflicts, but the whitespace cleanups (spaces => tabs) do > conflict, and I'd really like to start purging the spaces mess from KUT. > I'd based the v1 on [1], which is no longer required after your apic_ops percpu conversion series [2]. I've now based my series on yours and posted a v2 here [3]. [1] https://lore.kernel.org/kvm/20220406124002.13741-1-varad.gautam@suse.com/ [2] https://lore.kernel.org/all/20220121231852.1439917-1-seanjc@google.com/ [3] https://lore.kernel.org/kvm/20220412173407.13637-1-varad.gautam@suse.com/ > Paolo / Andrew, ping on my series, it still applies cleanly. > > [*] https://lore.kernel.org/all/20220121231852.1439917-1-seanjc@google.com >