Re: [Xen-devel] 回复: Re: [PATCH] Choose retpoline only when it is safe to use

On 2018/2/6 18:50, Andrew Cooper wrote: On 06/02/18 10:29, zhenzhong.duan wrote: 2018年2月6日 17:20于 Andrew Cooper <andrew.coop...@citrix.com <mailto:andrew.coop...@citrix.com>>写道： > > On 06/02/2018 09:13, Zhenzhong Duan wrote: > > 在 2018/2/6 16:59, Andrew Cooper 写道:

Re: [Xen-devel] 回复: Re: [PATCH] Choose retpoline only when it is safe to use

在 2018/2/6 19:56, Andrew Cooper 写道: On 06/02/18 11:41, Zhenzhong Duan wrote: On 2018/2/6 18:50, Andrew Cooper wrote: On 06/02/18 10:29, zhenzhong.duan wrote: 2018年2月6日 17:20于 Andrew Cooper <andrew.coop...@citrix.com <mailto:andrew.coop...@citrix.com>>写道： > > On 06/02/201

[Xen-devel] [PATCH] Fix a panic in SPEC_CTRL_ENTRY_FROM_INTR_IST

to cover the bug in certain case so we didn't reproduce without bti=0. Change to use %edx. Reviewed-by: Boris Ostrovsky <boris.ostrov...@oracle.com> Tested-by: Boris Ostrovsky <boris.ostrov...@oracle.com> Signed-off-by: Zhenzhong Duan <zhenzhong.d...@oracle.com> --- xen/include/asm-x

Re: [Xen-devel] [PATCH v2] x86: fix a crash in SPEC_CTRL_ENTRY_FROM_INTR_IST

. DO_OVERWRITE_RSB clobbers %eax and happend to cover the bug in certain case so we didn't reproduce without bti=0. Signed-off-by: Zhenzhong Duan <zhenzhong.d...@oracle.com> Re-do actual code change. Also drop an unused label. Signed-off-by: Jan Beulich <jbeul...@suse.com> --- a/xen/inc

Re: [Xen-devel] [PATCH v3] x86: fix a crash in SPEC_CTRL_ENTRY_FROM_INTR_IST

在 2018/2/14 17:58, Jan Beulich 写道: On 14.02.18 at 10:25, wrote: --- a/xen/include/asm-x86/spec_ctrl_asm.h +++ b/xen/include/asm-x86/spec_ctrl_asm.h @@ -269,28 +269,29 @@ * This is logical merge of DO_OVERWRITE_RSB and DO_SPEC_CTRL_ENTRY * maybexen=1, but with

[Xen-devel] [PATCH v3] x86: fix a crash in SPEC_CTRL_ENTRY_FROM_INTR_IST

to cover the bug in certain case so we didn't reproduce without bti=0. Change to use %edx. Also drop an unused label per Jan. Reported-by: Boris Ostrovsky <boris.ostrov...@oracle.com> Signed-off-by: Zhenzhong Duan <zhenzhong.d...@oracle.com> Signed-off-by: Jan Beulich <jbeul...@sus

Re: [Xen-devel] [PATCH] Fix a panic in SPEC_CTRL_ENTRY_FROM_INTR_IST

v...@oracle.com> Signed-off-by: Zhenzhong Duan <zhenzhong.d...@oracle.com> Two formal things: Please put tags in sequential (in time) order: reporter, author(s), reviewers/testers. And please follow patch submission rules - send them _to_ the list, with maintainers (and other intereste

[Xen-devel] [PATCH] Choose retpoline only when it is safe to use

When ( ibrs && thunk == THUNK_DEFAULT && !retpoline_safe() ) is true, thunk is set to THUNK_JMP rather than THUNK_RETPOLINE. When (!ibrs && thunk == THUNK_DEFAULT && !retpoline_safe() ) is true, we should do the same. Signed-off-by: Zhenzhong Duan <zhenzhong

Re: [Xen-devel] [PATCH] Choose retpoline only when it is safe to use

在 2018/2/6 16:59, Andrew Cooper 写道: On 06/02/2018 08:43, Zhenzhong Duan wrote: When ( ibrs && thunk == THUNK_DEFAULT && !retpoline_safe() ) is true, thunk is set to THUNK_JMP rather than THUNK_RETPOLINE. When (!ibrs && thunk == THUNK_DEFAULT && !retpolin

[Xen-devel] [PATCH] x86/mmcfg/drhd: Move acpi_mmcfg_init() before calling acpi_parse_dmar()

-by: Zhenzhong Duan Tested-by: Gopalasetty, Manoj --- xen/arch/x86/setup.c |8 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/xen/arch/x86/setup.c b/xen/arch/x86/setup.c index 8301de8..9af7426 100644 --- a/xen/arch/x86/setup.c +++ b/xen/arch/x86/setup.c @@ -1493,6 +1493,10

[Xen-devel] [PATCH] x86/mmcfg: Remove redundant code in pci_mmcfg_reject_broken()

No functional change. Signed-off-by: Zhenzhong Duan --- xen/arch/x86/x86_64/mmconfig-shared.c |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/xen/arch/x86/x86_64/mmconfig-shared.c b/xen/arch/x86/x86_64/mmconfig-shared.c index 7c3b7fd..4a6ca26 100644 --- a/xen/arch/x86

Re: [Xen-devel] [PATCH] x86/mmcfg/drhd: Move acpi_mmcfg_init() before calling acpi_parse_dmar()

On 2018/8/16 17:13, Zhenzhong Duan wrote: --- a/xen/arch/x86/setup.c +++ b/xen/arch/x86/setup.c @@ -1493,6 +1493,10 @@ void __init noreturn __start_xen(unsigned long mbi_p)   generic_apic_probe(); +    pt_pci_init(); + +    acpi_mmcfg_init(); +   acpi_boot_init

Re: [Xen-devel] [PATCH] x86/mmcfg/drhd: Move acpi_mmcfg_init() before calling acpi_parse_dmar()

On 2018/8/16 15:10, Jan Beulich wrote: On 16.08.18 at 07:10, wrote: On a multiple pci segment system such as HPE Superdome-Flex, pci config space from nonzero segment is accessed with mmcfg during acpi parsing DMAR region. First of all - can you please write a little more helpful (to

Re: [Xen-devel] [PATCH] x86/mmcfg/drhd: Move acpi_mmcfg_init() before calling acpi_parse_dmar()

在 2018/8/16 18:42, Jan Beulich 写道: On 16.08.18 at 11:30, wrote: On 2018/8/16 17:13, Zhenzhong Duan wrote: --- a/xen/arch/x86/setup.c +++ b/xen/arch/x86/setup.c @@ -1493,6 +1493,10 @@ void __init noreturn __start_xen(unsigned long mbi_p) generic_apic_probe(); +pt_pci_init

Re: [Xen-devel] [PATCH] x86/mmcfg/drhd: Move acpi_mmcfg_init() before calling acpi_parse_dmar()

在 2018/8/16 18:37, Jan Beulich 写道: On 16.08.18 at 11:13, wrote: On 2018/8/16 15:10, Jan Beulich wrote: Have you investigated the alternative of deferring acpi_dmar_init() to a later point, or at least the part of it that needs to do PCI config space accesses? I'm not currently convinced the

[Xen-devel] [PATCH v2 1/2] x86/mmcfg: Rename pt_pci_init() and call it in acpi_mmcfg_init()

Given what pt_pci_init() actually does, rename it properly and move its declaration to pci.h, move the only call in acpi_mmcfg_init(). No functional change. Signed-off-by: Zhenzhong Duan Tested-by: Gopalasetty, Manoj --- xen/arch/x86/setup.c |2 -- xen/arch/x86/x86_64

[Xen-devel] [PATCH v2 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

(). But acpi_parse_one_drhd(), acpi_parse_one_rmrr() and acpi_parse_one_atsr() all called acpi_parse_dev_scope() as their main job. Splitting those functions to two pieces looks less optimal and meaningless. Signed-off-by: Zhenzhong Duan Tested-by: Gopalasetty, Manoj --- xen/arch/x86/acpi/boot.c |2

[Xen-devel] [PATCH v3 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

-by: Zhenzhong Duan --- v3: fixup patch and description. xen/arch/x86/tboot.c |6 -- 1 files changed, 0 insertions(+), 6 deletions(-) diff --git a/xen/arch/x86/tboot.c b/xen/arch/x86/tboot.c index d5a5292..1006f95 100644 --- a/xen/arch/x86/tboot.c +++ b/xen/arch/x86/tboot.c @@ -461,8 +461,6

[Xen-devel] [PATCH v4 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

l years later. Some stale comments are also removed, No functional change. Signed-off-by: Zhenzhong Duan --- v4: fixup description per Jan xen/arch/x86/tboot.c |6 -- 1 files changed, 0 insertions(+), 6 deletions(-) diff --git a/xen/arch/x86/tboot.c b/xen/arch/x86/tboot.c index d5a

[Xen-devel] [PATCH v3 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

Dom0 boot has progressed far enough) if the range(s) isn't/aren't marked reserved in E820. Signed-off-by: Zhenzhong Duan Tested-by: Gopalasetty, Manoj --- xen/arch/x86/acpi/boot.c |2 ++ xen/arch/x86/setup.c |8 +++- 2 files changed, 5 insertions(+), 5 deletions(-) diff --git

Re: [Xen-devel] [PATCH v3 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

在 2018/8/22 16:42, Roger Pau Monné 写道: On Wed, Aug 22, 2018 at 04:39:05PM +0800, Zhenzhong Duan wrote: 在 2018/8/22 15:36, Roger Pau Monné 写道: On Tue, Aug 21, 2018 at 09:53:08PM -0700, Zhenzhong Duan wrote: pci_conf_read8() needs pci mmcfg mapping to work on multiple pci segments system

Re: [Xen-devel] [PATCH v3 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

在 2018/8/22 15:36, Roger Pau Monné 写道: On Tue, Aug 21, 2018 at 09:53:08PM -0700, Zhenzhong Duan wrote: pci_conf_read8() needs pci mmcfg mapping to work on multiple pci segments system such as HPE Superdome-Flex. Move acpi_mmcfg_init() call in acpi_boot_init() before calling acpi_parse_dmar

[Xen-devel] [PATCH RESEND v3 1/2] x86/mmcfg: Rename pt_pci_init() and call it in acpi_mmcfg_init()

Given what pt_pci_init() actually does, rename it properly and move its declaration to pci.h, move the only call in acpi_mmcfg_init(). No functional change. Signed-off-by: Zhenzhong Duan Tested-by: Gopalasetty, Manoj Acked-by: Jan Beulich --- xen/arch/x86/setup.c |2

[Xen-devel] [PATCH RESEND v3 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

Dom0 boot has progressed far enough) if the range(s) isn't/aren't marked reserved in E820. Signed-off-by: Zhenzhong Duan Tested-by: Gopalasetty, Manoj Reviewed-by: Roger Pau Monné --- xen/arch/x86/acpi/boot.c |2 ++ xen/arch/x86/setup.c |8 +++- 2 files changed, 5 insertions

[Xen-devel] 答复: [PATCH v4 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

parse_dmar(), which > 123c779379 ("VTd/dmar: Tweak how the DMAR table is clobbered") > added several years later. > > Some stale comments are also removed, No functional change. > > Signed-off-by: Zhenzhong Duan > --- > v4: fixup description per Jan > >

[Xen-devel] [PATCH 1/2] dmar: iommu mem leak fix

Release memory allocated for drhd iommu in error path. Signed-off-by: Zhenzhong Duan --- xen/drivers/passthrough/vtd/dmar.c |1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/xen/drivers/passthrough/vtd/dmar.c b/xen/drivers/passthrough/vtd/dmar.c index 46decd4..8c5fa80

[Xen-devel] [PATCH 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

heap table, fix the stale comments. Signed-off-by: Zhenzhong Duan --- xen/arch/x86/tboot.c |3 +-- xen/drivers/passthrough/vtd/dmar.c |2 ++ 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/xen/arch/x86/tboot.c b/xen/arch/x86/tboot.c index d5a5292..f22fa24

Re: [Xen-devel] [PATCH v2 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

On 2018/8/17 20:28, Jan Beulich wrote: On 17.08.18 at 09:01, wrote: pci_conf_read8() needs pci mmcfg mapping to work on multiple pci segments system such as HPE Superdome-Flex. Move acpi_mmcfg_init() call in acpi_boot_init() before calling acpi_parse_dmar() so that when pci_conf_read8() is

[Xen-devel] [PATCH v2 1/2] dmar: iommu mem leak fix

Release memory allocated for drhd iommu in error path. -v2: fixup wrong parameter hiden due to my removing -Werror Signed-off-by: Zhenzhong Duan --- xen/drivers/passthrough/vtd/dmar.c |1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/xen/drivers/passthrough/vtd/dmar.c b

[Xen-devel] [PATCH v2 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

heap table, fix the stale comments. This is osmetic change, it's unnecessory to zap a copy of DMAR table which is freed later. -v2: Add some description per Jan. Signed-off-by: Zhenzhong Duan --- xen/arch/x86/tboot.c |3 +-- xen/drivers/passthrough/vtd/dmar.c |2 ++ 2

[Xen-devel] 答复: [PATCH v2 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

which > is freed later. > > -v2: Add some description per Jan. > > Signed-off-by: Zhenzhong Duan > --- > xen/arch/x86/tboot.c |3 +-- > xen/drivers/passthrough/vtd/dmar.c |2 ++ > 2 files changed, 3 insertions(+), 2 deletions(-) > > diff -

Re: [Xen-devel] [PATCH 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

On 2018/8/20 16:44, Andrew Cooper wrote: On 20/08/2018 09:30, Jan Beulich wrote: On 20.08.18 at 05:32, wrote: When TBOOT enabled, acpi_parse_dmar() zap a copy of DMAR table rather than the real table, so make it controled by config option based on the fact that we already have done the real

Re: [Xen-devel] [PATCH 2/2] x86/dmar: zap DMAR signature for dom0 once in TBOOT case

comments. Signed-off-by: Zhenzhong Duan --- xen/arch/x86/tboot.c |3 +-- xen/drivers/passthrough/vtd/dmar.c |2 ++ 2 files changed, 3 insertions(+), 2 deletions(-) You've again failed to Cc maintainers (included now). Jan --- a/xen/arch/x86/tboot.c +++ b/xen/arch/x86

Re: [Xen-devel] [PATCH v2 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

On 2018/8/20 15:45, Jan Beulich wrote: On 20.08.18 at 05:38, wrote: On 2018/8/17 20:28, Jan Beulich wrote: On 17.08.18 at 09:01, wrote: pci_conf_read8() needs pci mmcfg mapping to work on multiple pci segments system such as HPE Superdome-Flex. Move acpi_mmcfg_init() call in

Re: [Xen-devel] [PATCH RESEND v3 1/2] x86/mmcfg: Rename pt_pci_init() and call it in acpi_mmcfg_init()

在 2018/8/27 16:16, Jan Beulich 写道: On 22.08.18 at 11:16, wrote: Given what pt_pci_init() actually does, rename it properly and move its declaration to pci.h, move the only call in acpi_mmcfg_init(). No functional change. Signed-off-by: Zhenzhong Duan Tested-by: Gopalasetty, Manoj Acked

[Xen-devel] [PATCH] x86/physdev: Remove redundant assignment in allocate_and_map_msi_pirq()

No functional change. Signed-off-by: Zhenzhong Duan --- xen/arch/x86/irq.c |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git a/xen/arch/x86/irq.c b/xen/arch/x86/irq.c index 87ef2e8..5253fd1 100644 --- a/xen/arch/x86/irq.c +++ b/xen/arch/x86/irq.c @@ -2701,8 +2701,6 @@ int

Re: [Xen-devel] [PATCH] x86/physdev: Remove redundant assignment in allocate_and_map_msi_pirq()

在 2018/7/20 18:13, Andrew Cooper 写道: On 20/07/18 11:13, Roger Pau Monné wrote: On Fri, Jul 20, 2018 at 02:29:34AM -0700, Zhenzhong Duan wrote: No functional change. Signed-off-by: Zhenzhong Duan --- xen/arch/x86/irq.c |2 -- 1 files changed, 0 insertions(+), 2 deletions(-) diff --git

[Xen-devel] [PATCH] xen/grant: Mute gcc warning in steal_linear_address()

Move reference of ol1e ahead or else we see below warning. cc1: warnings being treated as errors grant_table.c: In function 'replace_grant_pv_mapping': grant_table.c:142: warning: 'ol1e.l1' may be used uninitialized in this function Signed-off-by: Zhenzhong Duan --- xen/arch/x86/pv

[Xen-devel] 答复: [PATCH RESEND v3 2/2] x86/mmcfg/drhd: Move acpi_mmcfg_init() call before calling acpi_parse_dmar()

e precisely not at all until Dom0 boot has progressed far > enough) if the range(s) isn't/aren't marked reserved in E820. > > Signed-off-by: Zhenzhong Duan > Tested-by: Gopalasetty, Manoj > Reviewed-by: Roger Pau Monné > --- > xen/arch/x86/acpi/boot.c |2 ++ > xen/arc

Re: [Xen-devel] [PATCH] xen/grant: Mute gcc warning in steal_linear_address()

in this function Signed-off-by: Zhenzhong Duan I'm fine with the change, but going forward please be more specific with your descriptions: Whether or not warnings get emitted for a certain piece of code often highly depends on compiler version. Let us know which version you've run

Re: [Xen-devel] [PATCH] x86/boot: Disable IBRS in intr/nmi exit path at bootup stage

在 2018/3/26 21:39, Jan Beulich 写道: On 21.03.18 at 03:58, wrote: After reset, IBRS is disabled by processor, but a coming intr/nmi leave IBRS enabled after their exit. It's not necessory for bootup code to run in low performance with IBRS enabled. On ORACLE

[Xen-devel] [PATCH v2] x86/boot: Disable IBRS in intr/nmi exit path at bootup stage

rom sibling threads. v2: Use (system_state < SYS_STATE_active) to initialize use_shadow_spec_ctrl instead of literal 1 per Jan. Signed-off-by: Zhenzhong Duan <zhenzhong.d...@oracle.com> --- xen/include/asm-x86/spec_ctrl.h | 16 +++- 1 files changed, 15 insertions(+), 1 deleti

Re: [Xen-devel] [PATCH v2] x86/boot: Disable IBRS in intr/nmi exit path at bootup stage

On 2018/3/27 16:52, Jan Beulich wrote: On 27.03.18 at 06:52, wrote: After reset, IBRS is disabled by processor, but a coming intr/nmi leave IBRS enabled after their exit. It's not necessory for bootup code to run in low performance with IBRS enabled. On ORACLE

Re: [Xen-devel] [PATCH] x86/alt: Fix wrong usage of as_max in OLDINSTR_2

On 2018/3/28 13:52, Jan Beulich write: Zhenzhong Duan <zhenzhong.d...@oracle.com> 03/28/18 4:03 AM >>> When ALTERNATIVE_2 is used, we see below error during build. "error: macro "as_max" requires 2 arguments, but only 1 given" Signed-off-by: Zhenzhong Dua

[Xen-devel] [PATCH] x86/alt: Fix wrong usage of as_max in OLDINSTR_2

When ALTERNATIVE_2 is used, we see below error during build. "error: macro "as_max" requires 2 arguments, but only 1 given" Signed-off-by: Zhenzhong Duan <zhenzhong.d...@oracle.com> --- xen/include/asm-x86/alternative.h |4 ++-- 1 files changed, 2 insertions(+),

Re: [Xen-devel] [PATCH v2] x86/boot: Disable IBRS in intr/nmi exit path at bootup stage

On 2018/3/27 16:52, Jan Beulich wrote: On 27.03.18 at 06:52, wrote: After reset, IBRS is disabled by processor, but a coming intr/nmi leave IBRS enabled after their exit. It's not necessory for bootup code to run in low performance with IBRS enabled. On ORACLE

[Xen-devel] [PATCH] x86/boot: Disable IBRS in intr/nmi exit path at bootup stage

use_shadow_spec_ctrl with 1, IBRS is disabled in intr/nmi exit path at bootup stage. Then delay in construct_dom0 is ~50s. Signed-off-by: Zhenzhong Duan <zhenzhong.d...@oracle.com> --- xen/include/asm-x86/spec_ctrl.h |3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git

[Xen-devel] [PATCH v2 5/7] x86/xen: nopv parameter support for HVM guest

(), xen_pvh_domain() could be used for that purpose. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: xen-devel@lists.xenproject.org --- arch/x86/xen/enlighten_hvm.c | 18 -- 1

[Xen-devel] [PATCH v2 3/7] x86: Add nopv parameter to disable PV extensions

'xen_nopv' parameter for XEN platform but not for others. 'xen_nopv' can then be removed with this change. Suggested-by: Juergen Gross Signed-off-by: Zhenzhong Duan Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Jan Kiszka Cc: Boris Ostrovsky Cc: Stefano Stabellini Cc:

[Xen-devel] [PATCH v2 4/7] Revert "xen: Introduce 'xen_nopv' to disable PV extensions for HVM guests."

This reverts commit 8d693b911bb9c57009c24cb1772d205b84c7985c. Instead we use an unified parameter 'nopv' for all the hypervisor platforms. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc

[Xen-devel] [PATCH 3/6] x86: Add nopv parameter to disable PV extensions

it. There is already 'xen_nopv' parameter for XEN platform but not for others. 'xen_nopv' can then be removed with this change. Signed-off-by: Zhenzhong Duan Cc: xen-devel@lists.xenproject.org --- Documentation/admin-guide/kernel-parameters.txt | 4 arch/x86/kernel/cpu/hypervisor.c

[Xen-devel] [PATCH 4/6] Revert "xen: Introduce 'xen_nopv' to disable PV extensions for HVM guests."

This reverts commit 8d693b911bb9c57009c24cb1772d205b84c7985c. Instead we use an unified parameter 'nopv' for all the hypervisor platforms. Signed-off-by: Zhenzhong Duan Cc: xen-devel@lists.xenproject.org --- Documentation/admin-guide/kernel-parameters.txt | 4 arch/x86/xen

Re: [Xen-devel] [PATCH 3/6] x86: Add nopv parameter to disable PV extensions

On 2019/6/24 21:18, Juergen Gross wrote: On 23.06.19 15:01, Zhenzhong Duan wrote: In virtualization environment, PV extensions (drivers, interrupts, timers, etc) are enabled in the majority of use cases which is the best option. However, in some cases (kexec not fully working, benchmarking

Re: [Xen-devel] [PATCH v2 5/7] x86/xen: nopv parameter support for HVM guest

On 2019/6/25 20:31, Juergen Gross wrote: On 24.06.19 14:02, Zhenzhong Duan wrote: PVH guest needs PV extentions to work, so nopv parameter is ignored for PVH but not for HVM guest. In order for nopv parameter to take effect for HVM guest, we need to distinguish between PVH and HVM guest early

[Xen-devel] [PATCH v2] xen/pv: Fix a boot up hang revealed by int3 self test

we can fix it by using int3 and drop useless xenint3. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Andy Lutomirski Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov --- v2: fix up description. --- arch/x86/entry/entry_6

Re: [Xen-devel] [PATCH] xen/pv: Fix a boot up hang triggered by int3 self test

Sorry for the noise, it looks description is wrong. This is not a double pop, but xen pv taking the path with create_gap=0, I'll send a v2. Zhenzhong On 2019/7/11 12:47, Zhenzhong Duan wrote: Commit 7457c0da024b ("x86/alternatives: Add int3_emulate_call() selftest") reveals a bug

[Xen-devel] [PATCH v7 0/5] misc fixes to PV extensions code

Hi, In virtualization environment, PV extensions (drivers, interrupts, timers, etc) are enabled in the majority of use cases which is the best option. However, in some cases (kexec not fully working, benchmarking, etc) we want to disable PV extensions. We have xen_nopv for that purpose but only

[Xen-devel] [PATCH v7 2/5] x86: Add "nopv" parameter to disable PV extensions

r to struct hypervisor_x86 set to true for those guest types and call the detect functions only if nopv is false or ignore_nopv is true. Suggested-by: Juergen Gross Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Jan Kiszka Cc

[Xen-devel] [PATCH v7 4/5] x86/paravirt: Remove const mark from x86_hyper_xen_hvm variable

.. as "nopv" support needs it to be changeable at boot up stage. Checkpatch report warning, so move variable declarations from hypervisor.c to hypervisor.h Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo

Re: [Xen-devel] [PATCH v2] xen/pv: Fix a boot up hang revealed by int3 self test

On 2019/7/12 20:06, Peter Zijlstra wrote: On Thu, Jul 11, 2019 at 04:15:21PM +0800, Zhenzhong Duan wrote: diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index 4722ba2..2138d69 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -596,7 +596,7

Re: [Xen-devel] [PATCH v2] xen/pv: Fix a boot up hang revealed by int3 self test

On 2019/7/12 21:09, Peter Zijlstra wrote: On Fri, Jul 12, 2019 at 09:04:22PM +0800, Zhenzhong Duan wrote: On 2019/7/12 20:06, Peter Zijlstra wrote: On Thu, Jul 11, 2019 at 04:15:21PM +0800, Zhenzhong Duan wrote: diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index

[Xen-devel] [PATCH v7 5/5] x86/xen: Add "nopv" support for HVM guest

rmal boot entry same as HVM guest, it's hard to distinguish PVH and HVM guest at that time. In this case, we have to panic early if PVH is detected and nopv is enabled to avoid a worse situation later. Move xen_platform_hvm() after xen_hvm_guest_late_init() to avoid compile error. Signed-off-by

[Xen-devel] [PATCH v7 3/5] xen: Map "xen_nopv" parameter to "nopv" and mark it obsolete

Clean up unnecessory code after that operation. Signed-off-by: Zhenzhong Duan Reviewed-by: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov --- Documentation/admin-guide/kernel-parameters.txt | 2 ++ arch/x86/xen

[Xen-devel] [PATCH v7 1/5] x86/xen: Mark xen_hvm_need_lapic() and xen_x2apic_para_available() as __init

.. as they are only called at early bootup stage. In fact, other functions in x86_hyper_xen_hvm.init.* are all marked as __init. Unexport xen_hvm_need_lapic as it's never used outside. Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Boris Ostrovsky Cc: Stefano Stabellini Cc

[Xen-devel] [PATCH] xen/pv: Fix a boot up hang triggered by int3 self test

0x57a Fix it by removing xen_xenint3. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov --- arch/x86/include/asm/traps.h | 2 +- arch/x86/xen/enlighten_pv.c | 2 +- arch/x86/xen/xe

Re: [Xen-devel] [PATCH v6 4/4] x86/xen: Add "nopv" support for HVM guest

On 2019/7/8 21:46, Boris Ostrovsky wrote: On 7/7/19 5:15 AM, Zhenzhong Duan wrote: +static uint32_t __init xen_platform_hvm(void) +{ + if (xen_pv_domain()) + return 0; + + if (xen_pvh_domain() && nopv) { + /* Guest booting via the Xen-PVH bo

[Xen-devel] [PATCH v5 3/4] xen: Map "xen_nopv" parameter to "nopv" and mark it obsolete

Clean up unnecessory code after that operation. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov --- Documentation/admin-guide/kernel-parameters.txt | 2 ++ arch/x86/xen/enlighten_hvm.c

[Xen-devel] [PATCH v5 1/4] x86/xen: Mark xen_hvm_need_lapic() and xen_x2apic_para_available() as __init

.. as they are only called at early bootup stage. In fact, other functions in x86_hyper_xen_hvm.init.* are all marked as __init. Unexport xen_hvm_need_lapic as it's never used outside. Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Boris Ostrovsky Cc: Stefano Stabellini Cc

[Xen-devel] [PATCH v5 4/4] x86/xen: Add "nopv" support for HVM guest

rmal boot entry same as HVM guest, it's hard to distinguish PVH and HVM guest at that time. To handle that case, add a new function xen_hvm_nopv_guest_late_init() to detect PVH at a late time and panic itself if nopv enabled for a PVH guest. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky C

[Xen-devel] [PATCH v5 0/4] misc fixes to PV extentions code

Hi, In virtualization environment, PV extensions (drivers, interrupts, timers, etc) are enabled in the majority of use cases which is the best option. However, in some cases (kexec not fully working, benchmarking, etc) we want to disable PV extensions. We have xen_nopv for that purpose but only

[Xen-devel] [PATCH v5 2/4] x86: Add "nopv" parameter to disable PV extensions

r to struct hypervisor_x86 set to true for those guest types and call the detect functions only if nopv is false or ignore_nopv is true. Suggested-by: Juergen Gross Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Jan Kiszka Cc

Re: [Xen-devel] [PATCH v4 5/5] xen: Add 'xen_nopv' parameter back for backward compatibility

On 2019/7/3 2:13, Boris Ostrovsky wrote: On 7/1/19 1:19 AM, Zhenzhong Duan wrote: Map 'xen_nopv' to 'nopv' and mark 'xen_nopv' obsolete in kernel-parameters.txt I am not sure we want patch #3, why not do everything in a single patch? Thanks for review. I'll fix all those based on your

Re: [Xen-devel] [PATCH v5 4/4] x86/xen: Add "nopv" support for HVM guest

On 2019/7/5 21:06, Boris Ostrovsky wrote: On 7/2/19 9:19 PM, Zhenzhong Duan wrote: PVH guest needs PV extentions to work, so "nopv" parameter should be ignored for PVH but not for HVM guest. If PVH guest boots up via the Xen-PVH boot entry, xen_pvh is set early, we know it's

[Xen-devel] [PATCH v6 4/4] x86/xen: Add "nopv" support for HVM guest

rmal boot entry same as HVM guest, it's hard to distinguish PVH and HVM guest at that time. In this case, we have to panic early if PVH is detected and nopv is enabled to avoid a worse situation later. Move xen_platform_hvm() after xen_hvm_guest_late_init() to avoid compile error. Signed-off-by

[Xen-devel] [PATCH v6 1/4] x86/xen: Mark xen_hvm_need_lapic() and xen_x2apic_para_available() as __init

.. as they are only called at early bootup stage. In fact, other functions in x86_hyper_xen_hvm.init.* are all marked as __init. Unexport xen_hvm_need_lapic as it's never used outside. Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Boris Ostrovsky Cc: Stefano Stabellini Cc

[Xen-devel] [PATCH v6 3/4] xen: Map "xen_nopv" parameter to "nopv" and mark it obsolete

Clean up unnecessory code after that operation. Signed-off-by: Zhenzhong Duan Reviewed-by: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov --- Documentation/admin-guide/kernel-parameters.txt | 2 ++ arch/x86/xen

[Xen-devel] [PATCH v6 0/4] misc fixes to PV extensions code

Hi, In virtualization environment, PV extensions (drivers, interrupts, timers, etc) are enabled in the majority of use cases which is the best option. However, in some cases (kexec not fully working, benchmarking, etc) we want to disable PV extensions. We have xen_nopv for that purpose but only

[Xen-devel] [PATCH v6 2/4] x86: Add "nopv" parameter to disable PV extensions

r to struct hypervisor_x86 set to true for those guest types and call the detect functions only if nopv is false or ignore_nopv is true. Suggested-by: Juergen Gross Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Jan Kiszka Cc

[Xen-devel] [PATCH v3] xen/pv: Fix a boot up hang revealed by int3 self test

te a gap, we can fix it by using int3 and drop useless xenint3. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Andy Lutomirski Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Andrew Cooper --- bootup test pass with

Re: [Xen-devel] [PATCH v2] xen/pv: Fix a boot up hang revealed by int3 self test

On 2019/7/12 22:06, Andrew Cooper wrote: On 11/07/2019 03:15, Zhenzhong Duan wrote: Commit 7457c0da024b ("x86/alternatives: Add int3_emulate_call() selftest") is used to ensure there is a gap setup in exception stack which could be used for inserting call return address. This gap

[Xen-devel] [PATCH v3 4/4] x86/xen: Add 'nopv' support for HVM guest

as HVM guest, it's hard to distinguish PVH and HVM guest at that time. To handle that case, add a new function xen_hvm_nopv_guest_late_init() to detect PVH at a late time and panic itself if 'nopv' enabled for a PVH guest. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc

[Xen-devel] [PATCH v3 1/4] x86/xen: Mark xen_hvm_need_lapic() and xen_x2apic_para_available() as __init

.. as they are only called at early bootup stage. In fact, other functions in x86_hyper_xen_hvm.init.* are all marked as __init. Unexport xen_hvm_need_lapic as it's never used outside. Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Boris Ostrovsky Cc: Stefano Stabellini Cc

[Xen-devel] [PATCH v4 4/5] x86/xen: Add 'nopv' support for HVM guest

as HVM guest, it's hard to distinguish PVH and HVM guest at that time. To handle that case, add a new function xen_hvm_nopv_guest_late_init() to detect PVH at a late time and panic itself if 'nopv' enabled for a PVH guest. Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc

[Xen-devel] [PATCH v4 1/5] x86/xen: Mark xen_hvm_need_lapic() and xen_x2apic_para_available() as __init

.. as they are only called at early bootup stage. In fact, other functions in x86_hyper_xen_hvm.init.* are all marked as __init. Unexport xen_hvm_need_lapic as it's never used outside. Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Boris Ostrovsky Cc: Stefano Stabellini Cc

[Xen-devel] [PATCH v4 2/5] x86: Add nopv parameter to disable PV extensions

'xen_nopv' parameter for XEN platform but not for others. 'xen_nopv' can then be removed with this change. Suggested-by: Juergen Gross Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Jan Kiszka Cc: Boris Ostrovsky C

[Xen-devel] [PATCH v4 3/5] Revert "xen: Introduce 'xen_nopv' to disable PV extensions for HVM guests."

This reverts commit 8d693b911bb9c57009c24cb1772d205b84c7985c. Instead we use an unified parameter 'nopv' for all the hypervisor platforms. Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo

[Xen-devel] [PATCH v4 0/5] misc fixes to PV extentions code

Hi, In virtualization environment, PV extensions (drivers, interrupts, timers, etc) are enabled in the majority of use cases which is the best option. However, in some cases (kexec not fully working, benchmarking, etc) we want to disable PV extensions. We have xen_nopv for that purpose but only

[Xen-devel] [PATCH v4 5/5] xen: Add 'xen_nopv' parameter back for backward compatibility

Map 'xen_nopv' to 'nopv' and mark 'xen_nopv' obsolete in kernel-parameters.txt Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov --- Documentation/admin-guide/kernel-parameters.txt | 6

[Xen-devel] [PATCH v3 2/4] x86: Add nopv parameter to disable PV extensions

'xen_nopv' parameter for XEN platform but not for others. 'xen_nopv' can then be removed with this change. Suggested-by: Juergen Gross Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Jan Kiszka Cc: Boris Ostrovsky C

[Xen-devel] [PATCH v3 3/4] Revert "xen: Introduce 'xen_nopv' to disable PV extensions for HVM guests."

This reverts commit 8d693b911bb9c57009c24cb1772d205b84c7985c. Instead we use an unified parameter 'nopv' for all the hypervisor platforms. Signed-off-by: Zhenzhong Duan Reviewed-by: Juergen Gross Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo

[Xen-devel] [PATCH v3 0/4] misc fixes to PV extentions code

Hi, In virtualization environment, PV extensions (drivers, interrupts, timers, etc) are enabled in the majority of use cases which is the best option. However, in some cases (kexec not fully working, benchmarking, etc) we want to disable PV extensions. We have xen_nopv for that purpose but only

Re: [Xen-devel] [PATCH v3 3/4] Revert "xen: Introduce 'xen_nopv' to disable PV extensions for HVM guests."

On 2019/7/2 11:48, Boris Ostrovsky wrote: On Mon, Jul 01, 2019 at 10:20:27AM +0800, Zhenzhong Duan wrote: This reverts commit 8d693b911bb9c57009c24cb1772d205b84c7985c. Instead we use an unified parameter 'nopv' for all the hypervisor platforms. Signed-off-by: Zhenzhong Duan Reviewed

Re: [Xen-devel] [PATCH v7 4/5] x86/paravirt: Remove const mark from x86_hyper_xen_hvm variable

On 2019/7/16 18:57, Juergen Gross wrote: On 11.07.19 14:02, Zhenzhong Duan wrote: .. as "nopv" support needs it to be changeable at boot up stage. Checkpatch report warning, so move variable declarations from hypervisor.c to hypervisor.h Signed-off-by: Zhenzhong Duan Cc: Boris Ost

[Xen-devel] [PATCH v8 4/5] x86/paravirt: Remove const mark from x86_hyper_xen_hvm variable

.. as "nopv" support needs it to be changeable at boot up stage. Checkpatch reports warning, so move variable declarations from hypervisor.c to hypervisor.h Signed-off-by: Zhenzhong Duan Cc: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo

[Xen-devel] [PATCH v8 5/5] x86/xen: Add "nopv" support for HVM guest

platform_hvm() after xen_hvm_guest_late_init() to avoid compile error. Signed-off-by: Zhenzhong Duan Reviewed-by: Boris Ostrovsky Cc: Juergen Gross Cc: Stefano Stabellini Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov --- arch/x86/include/asm/x86_init.h | 2 ++ arch/x86/kernel/

Re: [Xen-devel] [PATCH v6 4/4] x86/xen: Add "nopv" support for HVM guest

On 2019/7/9 22:54, Boris Ostrovsky wrote: On 7/9/19 12:20 AM, Zhenzhong Duan wrote: -const __initconst struct hypervisor_x86 x86_hyper_xen_hvm = { +static uint32_t __init xen_platform_hvm(void) +{ +   uint32_t xen_domain = xen_cpuid_base(); +   struct x86_hyper_init *h