Message ID | cover.1415926876.git.geoff@infradead.org |
---|---|
State | New |
Headers | show |
Hello. On 11/14/2014 4:22 AM, Geoff Levand wrote: > Add runtime checks that fail the arm64 kexec syscall for situations that would > result in system instability do to problems in the KVM and EFI kernel support. > These checks should be removed when the KVM and EFI problems are fixed. > Signed-off-by: Geoff Levand <geoff@infradead.org> > --- > arch/arm64/kernel/machine_kexec.c | 22 ++++++++++++++++++++++ > 1 file changed, 22 insertions(+) > diff --git a/arch/arm64/kernel/machine_kexec.c b/arch/arm64/kernel/machine_kexec.c > index 775bcc3..5b62bd1f 100644 > --- a/arch/arm64/kernel/machine_kexec.c > +++ b/arch/arm64/kernel/machine_kexec.c > @@ -16,6 +16,12 @@ > #include <asm/cacheflush.h> > #include <asm/system_misc.h> > +/* TODO: Remove this include when KVM can support a kexc reboot. */ s/kexc/kexec/ here and below. > +#include <asm/virt.h> > + > +/* TODO: Remove this include when EFI can support a kexc reboot. */ > +#include <linux/efi.h> > + > /* Global variables for the relocate_kernel routine. */ > extern const unsigned char relocate_new_kernel[]; > extern const unsigned long relocate_new_kernel_size; > @@ -101,6 +107,22 @@ int machine_kexec_prepare(struct kimage *image) > > kexec_image_info(image); > > + /* TODO: Remove this message when KVM can support a kexc reboot. */ > + if (IS_ENABLED(CONFIG_KVM) && is_hyp_mode_available()) { > + pr_err("%s: Your kernel is configued with KVM support " Configured. > + "(CONFIG_KVM=y) which currently does not allow for " > + "kexec re-boot.\n", __func__); > + return -ENOSYS; > + } > + > + /* TODO: Remove this message when EFI can support a kexc reboot. */ > + if (efi_enabled(EFI_BOOT)) { > + pr_err("%s: Your kernel is using the Extensible Firmware " > + "Interface (EFI) which currently does not allow for " > + "kexec re-boot.\n", __func__); > + return -ENOSYS; > + } > + > return 0; > } WBR, Sergei
On 11/14/2014 09:28 PM, Geoff Levand wrote: > Add runtime checks that fail the arm64 kexec syscall for situations that would > result in system instability do to problems in the KVM and EFI kernel support. > These checks should be removed when the KVM and EFI problems are fixed. > Signed-off-by: Geoff Levand <geoff@infradead.org> > --- > v2: Typo fixes. > arch/arm64/kernel/machine_kexec.c | 22 ++++++++++++++++++++++ > 1 file changed, 22 insertions(+) > diff --git a/arch/arm64/kernel/machine_kexec.c b/arch/arm64/kernel/machine_kexec.c > index 775bcc3..faeb4f1 100644 > --- a/arch/arm64/kernel/machine_kexec.c > +++ b/arch/arm64/kernel/machine_kexec.c [...] > @@ -101,6 +107,22 @@ int machine_kexec_prepare(struct kimage *image) > > kexec_image_info(image); > > + /* TODO: Remove this message when KVM can support a kexec reboot. */ > + if (IS_ENABLED(CONFIG_KVM) && is_hyp_mode_available()) { > + pr_err("%s: Your kernel is configured with KVM support " > + "(CONFIG_KVM=y) which currently does not allow for " > + "kexec re-boot.\n", __func__); Er, failed to notice the first time: you do not need to break up the message. This helps finding the messages in the kernel code. If you would have run the patch thru scripst/checkpatch.pl you would have gotten the actual warnings about that... [...] WBR, Sergei
Geoff, Is there any Linaro card for the KVM soft restart issue? Ccing Andrew Jones since he might be interested in KVM issues. On 11/14/14 at 01:22am, Geoff Levand wrote: > Hi All, > > This series adds the core support for kexec re-boots on arm64. I have tested > with the ARM VE fast model, the ARM Base model and the ARM Foundation > model with various kernel config options for both the first and second stage > kernels. > > To load a second stage kernel and execute a kexec re-boot on arm64 my patches to > kexec-tools [2], which have not yet been merged upstream, are needed. > > Patches 1-3 rework the arm64 hcall mechanism to give the arm64 soft_restart() > routine the ability to switch exception levels from EL1 to EL2 for kernels that > were entered in EL2. > > Patch 4 moves proc-macros.S from arm64/mm to arm64/include/asm so that the > dcache_line_size macro it defines can be uesd by kexec's relocate kernel > routine. > > Patches 5-8 add the actual kexec support. > > Please consider all patches for inclusion. Any comments or suggestions on how > to improve are welcome. > > [1] https://git.linaro.org/people/geoff.levand/linux-kexec.git > [2] https://git.linaro.org/people/geoff.levand/kexec-tools.git > > Several things are known to have problems on kexec re-boot: > > spin-table > ---------- > > PROBLEM: The spin-table enable method does not implement all the methods needed > for CPU hot-plug, so the first stage kernel cannot be shutdown properly. > > WORK-AROUND: Upgrade to system firmware that provides PSCI enable method > support, OR build the first stage kernel with CONFIG_SMP=n, OR pass 'maxcpus=1' > on the first stage kernel command line. > > FIX: Upgrade system firmware to provide PSCI enable method support or add > missing spin-table support to the kernel. > > KVM > --- > > PROBLEM: KVM acquires hypervisor resources on startup, but does not free those > resources on shutdown, so the first stage kernel cannot be shutdown properly. > > WORK-AROUND: Build the first stage kernel with CONFIG_KVM=n. > > FIX: Fix KVM to support soft_restart(). KVM needs to restore default exception > vectors, etc. > > UEFI > ---- > > PROBLEM: UEFI does not manage its runtime services virtual mappings in a way > that is compatible with a kexec re-boot, so the second stage kernel hangs on > boot-up. > > WORK-AROUND: Disable UEFI in firmware, OR use EFI uboot emulation if provided > by your firmware. > > FIX: Fix kernel to manage UEFI virtual mappings properly. > > /memreserve/ > ---------- > > PROBLEM: The use of device tree /memreserve/ entries is not compatible with > kexec re-boot. The second stage kernel will use the reserved regions and the > system will become unstable. > > WORK-AROUND: Pass a user specified DTB using the kexec --dtb option. > > FIX: Convert device tree source files (.dts) and/or bootloaders to use device > tree reserved-memory nodes to specify reserved memory or provide an in kernel > mechanism to export the /memreserve/ entries to user space. > > ACPI > ---- > > PROBLEM: The kernel for ACPI based systems does not export a device tree to the > standard user space location of 'proc/device-tree'. Current applications > expect to access device tree information from this standard location. > > WORK-AROUND: Disable ACPI in firmware, OR pass 'acpi=off' on the first stage > kernel command line, OR pass a user specified DTB using the kexec --dtb option. > > FIX: Fix ACPI kernel support to export device tree information suitable for a > kexec re-boot, and if this information is exported to somewhere other than > 'proc/device-tree' update existing kexec user applications to use this new > interface. > > ---------------------------------------------------------------- > The following changes since commit 302cd37c417d59549d073a8986fd028998409cb9: > > arm64: kvm: eliminate literal pool entries (2014-11-07 15:04:54 +0000) > > are available in the git repository at: > > git://git.linaro.org/people/geoff.levand/linux-kexec.git kexec-v6 > > for you to fetch changes up to 3fc5a4ba22342ccd8e7e5a09620b00f4b81667a6: > > arm64/kexec: Enable kexec in the arm64 defconfig (2014-11-13 16:52:16 -0800) > > ---------------------------------------------------------------- > Geoff Levand (8): > arm64: Convert hcalls to use ISS field > arm64: Add new hcall HVC_CALL_FUNC > arm64: Add EL2 switch to soft_restart > arm64: Move proc-macros.S to include/asm > arm64/kexec: Add core kexec support > arm64/kexec: Add pr_devel output > arm64/kexec: Add checks for KVM and EFI > arm64/kexec: Enable kexec in the arm64 defconfig > > arch/arm64/Kconfig | 9 ++ > arch/arm64/configs/defconfig | 1 + > arch/arm64/include/asm/kexec.h | 47 ++++++ > arch/arm64/include/asm/proc-fns.h | 4 +- > arch/arm64/{mm => include/asm}/proc-macros.S | 0 > arch/arm64/include/asm/virt.h | 33 ++++ > arch/arm64/kernel/Makefile | 1 + > arch/arm64/kernel/hyp-stub.S | 45 ++++-- > arch/arm64/kernel/machine_kexec.c | 232 +++++++++++++++++++++++++++ > arch/arm64/kernel/process.c | 10 +- > arch/arm64/kernel/relocate_kernel.S | 160 ++++++++++++++++++ > arch/arm64/kvm/hyp.S | 18 ++- > arch/arm64/mm/cache.S | 3 +- > arch/arm64/mm/proc.S | 50 ++++-- > include/uapi/linux/kexec.h | 1 + > 15 files changed, 576 insertions(+), 38 deletions(-) > create mode 100644 arch/arm64/include/asm/kexec.h > rename arch/arm64/{mm => include/asm}/proc-macros.S (100%) > create mode 100644 arch/arm64/kernel/machine_kexec.c > create mode 100644 arch/arm64/kernel/relocate_kernel.S > > -- > 1.9.1 > > > _______________________________________________ > kexec mailing list > kexec@lists.infradead.org > http://lists.infradead.org/mailman/listinfo/kexec > >
On 11/27/14 at 02:19pm, Dave Young wrote: > Geoff, > > Is there any Linaro card for the KVM soft restart issue? > > Ccing Andrew Jones since he might be interested in KVM issues. > > On 11/14/14 at 01:22am, Geoff Levand wrote: > > Hi All, > > > > This series adds the core support for kexec re-boots on arm64. I have tested > > with the ARM VE fast model, the ARM Base model and the ARM Foundation > > model with various kernel config options for both the first and second stage > > kernels. > > > > To load a second stage kernel and execute a kexec re-boot on arm64 my patches to > > kexec-tools [2], which have not yet been merged upstream, are needed. > > > > Patches 1-3 rework the arm64 hcall mechanism to give the arm64 soft_restart() > > routine the ability to switch exception levels from EL1 to EL2 for kernels that > > were entered in EL2. > > > > Patch 4 moves proc-macros.S from arm64/mm to arm64/include/asm so that the > > dcache_line_size macro it defines can be uesd by kexec's relocate kernel > > routine. > > > > Patches 5-8 add the actual kexec support. > > > > Please consider all patches for inclusion. Any comments or suggestions on how > > to improve are welcome. > > > > [1] https://git.linaro.org/people/geoff.levand/linux-kexec.git > > [2] https://git.linaro.org/people/geoff.levand/kexec-tools.git > > > > Several things are known to have problems on kexec re-boot: > > > > spin-table > > ---------- > > > > PROBLEM: The spin-table enable method does not implement all the methods needed > > for CPU hot-plug, so the first stage kernel cannot be shutdown properly. > > > > WORK-AROUND: Upgrade to system firmware that provides PSCI enable method > > support, OR build the first stage kernel with CONFIG_SMP=n, OR pass 'maxcpus=1' > > on the first stage kernel command line. > > > > FIX: Upgrade system firmware to provide PSCI enable method support or add > > missing spin-table support to the kernel. > > > > KVM > > --- > > > > PROBLEM: KVM acquires hypervisor resources on startup, but does not free those > > resources on shutdown, so the first stage kernel cannot be shutdown properly. > > > > WORK-AROUND: Build the first stage kernel with CONFIG_KVM=n. > > > > FIX: Fix KVM to support soft_restart(). KVM needs to restore default exception > > vectors, etc. > > > > UEFI > > ---- > > > > PROBLEM: UEFI does not manage its runtime services virtual mappings in a way > > that is compatible with a kexec re-boot, so the second stage kernel hangs on > > boot-up. > > > > WORK-AROUND: Disable UEFI in firmware, OR use EFI uboot emulation if provided > > by your firmware. > > > > FIX: Fix kernel to manage UEFI virtual mappings properly. > > > > /memreserve/ > > ---------- > > > > PROBLEM: The use of device tree /memreserve/ entries is not compatible with > > kexec re-boot. The second stage kernel will use the reserved regions and the > > system will become unstable. > > > > WORK-AROUND: Pass a user specified DTB using the kexec --dtb option. > > > > FIX: Convert device tree source files (.dts) and/or bootloaders to use device > > tree reserved-memory nodes to specify reserved memory or provide an in kernel > > mechanism to export the /memreserve/ entries to user space. > > > > ACPI > > ---- > > > > PROBLEM: The kernel for ACPI based systems does not export a device tree to the > > standard user space location of 'proc/device-tree'. Current applications > > expect to access device tree information from this standard location. > > > > WORK-AROUND: Disable ACPI in firmware, OR pass 'acpi=off' on the first stage > > kernel command line, OR pass a user specified DTB using the kexec --dtb option. > > > > FIX: Fix ACPI kernel support to export device tree information suitable for a > > kexec re-boot, and if this information is exported to somewhere other than > > 'proc/device-tree' update existing kexec user applications to use this new > > interface. > > > > ---------------------------------------------------------------- > > The following changes since commit 302cd37c417d59549d073a8986fd028998409cb9: > > > > arm64: kvm: eliminate literal pool entries (2014-11-07 15:04:54 +0000) > > > > are available in the git repository at: > > > > git://git.linaro.org/people/geoff.levand/linux-kexec.git kexec-v6 > > Tested your master branch, works for me with below setup: UEFI boot, CONFIG_KVM=n, maxcpus=1, --lite But --dtb still does not work if there's no /proc/device-tree, failed as below: unrecoverable error: could not scan "/proc/device-tree/": No such file or directory. kexec without --lite also does not work, looks like it hangs after "Bye" .. Thanks Dave
Hi Dave,
On Thu, 2014-11-27 at 14:19 +0800, Dave Young wrote:
> Is there any Linaro card for the KVM soft restart issue?
https://cards.linaro.org/browse/KWG-611
I have a skeleton patch for the fix that needs to be filled
out in both in the Linaro card and in the commit (arm64/kvm:
Add KVM cpu shutdown) my master branch here:
http://git.linaro.org/people/geoff.levand/linux-kexec.git
-Geoff
Hi Dave, On Fri, 2014-11-28 at 14:56 +0800, Dave Young wrote: > Tested your master branch, works for me with below setup: > UEFI boot, CONFIG_KVM=n, maxcpus=1, --lite > > But --dtb still does not work if there's no /proc/device-tree, failed as below: > unrecoverable error: could not scan "/proc/device-tree/": No such file or directory. > > kexec without --lite also does not work, looks like it hangs after "Bye" .. I'll be away from the office this week. I'll look into this when I return. -Geoff