[Xen-devel] [ovmf test] 99968: all pass - PUSHED

flight 99968 ovmf real [real] http://logs.test-lab.xenproject.org/osstest/logs/99968/ Perfect :-) All tests in this flight passed as required version targeted for testing: ovmf 0667e985270b5c2cc5d89e66bcff7e5804eee21f baseline version: ovmf

[Xen-devel] [xen-4.5-testing test] 99963: regressions - FAIL

flight 99963 xen-4.5-testing real [real] http://logs.test-lab.xenproject.org/osstest/logs/99963/ Regressions :-( Tests which did not succeed and are blocking, including tests which could not be run: test-amd64-amd64-xl-pvh-amd 6 xen-boot fail REGR. vs. 99752

[Xen-devel] [xen-4.6-testing test] 99962: regressions - FAIL

flight 99962 xen-4.6-testing real [real] http://logs.test-lab.xenproject.org/osstest/logs/99962/ Regressions :-( Tests which did not succeed and are blocking, including tests which could not be run: test-amd64-amd64-xl-pvh-intel 6 xen-boot fail REGR. vs. 99902

[Xen-devel] [xen-unstable test] 99959: regressions - FAIL

flight 99959 xen-unstable real [real] http://logs.test-lab.xenproject.org/osstest/logs/99959/ Regressions :-( Tests which did not succeed and are blocking, including tests which could not be run: test-armhf-armhf-libvirt-qcow2 9 debian-di-install fail REGR. vs. 99917 test-armhf-armhf-xl

[Xen-devel] [PATCH v4 12/19] efi: introduce EFI_RS to ease control on runtime services usage

Suggested-by: Jan Beulich Signed-off-by: Daniel Kiper --- xen/arch/x86/domain_page.c |2 +- xen/arch/x86/shutdown.c|2 +- xen/arch/x86/time.c|2 +- xen/common/efi/boot.c |4 xen/include/xen/efi.h |1 + 5

[Xen-devel] [PATCH v4 15/19] x86/efi: create new early memory allocator

There is a problem with place_string() which is used as early memory allocator. It gets memory chunks starting from start symbol and goes down. Sadly this does not work when Xen is loaded using multiboot2 protocol because then the start lives on 1 MiB address and we should not allocate a memory

[Xen-devel] [PATCH v4 19/19] x86: add multiboot2 protocol support for relocatable images

Add multiboot2 protocol support for relocatable images. Only GRUB2 with "multiboot2: Add support for relocatable images" patch understands that feature. Older multiboot protocol (regardless of version) compatible loaders ignore it and everything works as usual. Signed-off-by: Daniel Kiper

[Xen-devel] [PATCH v4 17/19] x86/boot: implement early command line parser in C

Current early command line parser implementation in assembler is very difficult to change to relocatable stuff using segment registers. This requires a lot of changes in very weird and fragile code. So, reimplement this functionality in C. This way code will be relocatable out of the box (without

[Xen-devel] [PATCH v4 18/19] x86: make Xen early boot code relocatable

Every multiboot protocol (regardless of version) compatible image must specify its load address (in ELF or multiboot header). Multiboot protocol compatible loader have to load image at specified address. However, there is no guarantee that the requested memory region (in case of Xen it starts at 1

[Xen-devel] [PATCH v4 14/19] efi: build xen.gz with EFI code

Build xen.gz with EFI code. We need this to support multiboot2 protocol on EFI platforms. If we wish to load non-ELF file using multiboot (v1) or multiboot2 then it must contain "linear" (or "flat") representation of code and data. This is requirement of both boot protocols. Currently, PE file

[Xen-devel] [PATCH v4 16/19] x86: add multiboot2 protocol support for EFI platforms

This way Xen can be loaded on EFI platforms using GRUB2 and other boot loaders which support multiboot2 protocol. Signed-off-by: Daniel Kiper --- v4 - suggestions/fixes: - remove redundant BSS alignment, - update BSS alignment check, - use __set_bit() instead of

[Xen-devel] [PATCH v4 13/19] efi: EFI_RS bit in efi.flags must be controlled by efi=[no-]rs command line argument

Otherwise efi_enabled(EFI_RS) check is unreliable. Signed-off-by: Daniel Kiper --- xen/common/efi/boot.c | 10 ++ 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/xen/common/efi/boot.c b/xen/common/efi/boot.c index dd6b0a8..95cb25f 100644 ---

[Xen-devel] [PATCH v4 09/19] x86: add multiboot2 protocol support

Add multiboot2 protocol support. Alter min memory limit handling as we now may not find it from either multiboot (v1) or multiboot2. This way we are laying the foundation for EFI + GRUB2 + Xen development. Signed-off-by: Daniel Kiper --- v4 - suggestions/fixes: -

[Xen-devel] [PATCH v4 08/19] x86/boot/reloc: Rename some variables and rearrange code a bit

Replace mbi with mbi_out and mbi_old with mbi_in and rearrange code a bit to make it more readable. Additionally, this way multiboot (v1) protocol implementation and future multiboot2 protocol implementation will use the same variable naming convention. Signed-off-by: Daniel Kiper

[Xen-devel] [PATCH v4 03/19] x86/boot: create *.lnk files with linker script

Newer GCC (e.g. gcc version 5.1.1 20150618 (Red Hat 5.1.1-4) (GCC)) does some code optimizations by creating data sections (e.g. jump addresses for C switch/case are calculated using data in .rodata section). This thing is not accepted by *.lnk build recipe which requires that only .text section

[Xen-devel] [PATCH v4 04/19] x86/boot/reloc: reduce assembly usage as much as possible

Next patch will leave just required jmp instruction in xen/x86/boot/reloc.c. Signed-off-by: Daniel Kiper --- xen/arch/x86/boot/build32.lds |1 + xen/arch/x86/boot/build32.mk |2 +- xen/arch/x86/boot/reloc.c | 52 -

[Xen-devel] [PATCH v4 11/19] efi: create efi_enabled()

First of all we need to differentiate between legacy BIOS and EFI platforms during runtime, not during build, because one image will have legacy and EFI code and can be executed on both platforms. Additionally, we need more fine grained knowledge about EFI environment and check for EFI platform

[Xen-devel] [PATCH v4 10/19] efi: move efi struct initialization to xen/common/lib.c

A subsequent patch adds efi struct flags member which is used during runtime to differentiate between legacy BIOS and EFI platforms and multiboot2 and EFI native loader. So, efi symbol have to proper representation in ELF and PE Xen image. Hence, move efi struct initialization to xen/common/lib.c

[Xen-devel] [PATCH v4 06/19] x86/boot/reloc: create generic alloc and copy functions

Create generic alloc and copy functions. We need separate tools for memory allocation and copy to provide multiboot2 protocol support. Signed-off-by: Daniel Kiper --- v4 - suggestions/fixes: - avoid assembly usage. v3 - suggestions/fixes: - use "g" constraint

[Xen-devel] [PATCH v4 01/19] x86: allow EFI reboot method neither on EFI platforms...

..nor EFI platforms with runtime services disabled. Suggested-by: Jan Beulich Signed-off-by: Daniel Kiper --- xen/arch/x86/shutdown.c |3 +++ 1 file changed, 3 insertions(+) diff --git a/xen/arch/x86/shutdown.c b/xen/arch/x86/shutdown.c index

[Xen-devel] [PATCH v4 07/19] x86/boot: use %ecx instead of %eax

Use %ecx instead of %eax to store low memory upper limit from EBDA. This way we do not wipe multiboot protocol identifier. It is needed in reloc() to differentiate between multiboot (v1) and multiboot2 protocol. Signed-off-by: Daniel Kiper Reviewed-by: Andrew Cooper

[Xen-devel] [PATCH v4 05/19] x86/boot: call reloc() using stdcall calling convention

Current reloc() call method makes confusion and does not scale well for more arguments. And patch adding multiboot2 protocol support have to pass 3 arguments instead of 2. Hence, move reloc() call to stdcall calling convention. This way, in comparison to cdecl calling convention, we do not need to

[Xen-devel] [PATCH v4 00/19] x86: multiboot2 protocol support

Hi, I am sending fourth version of multiboot2 protocol support for legacy BIOS and EFI platforms. This patch series release contains fixes for all known issues. The final goal is xen.efi binary file which could be loaded by EFI loader, multiboot (v1) protocol (only on legacy BIOS platforms) and

[Xen-devel] [PATCH v4 02/19] x86/boot: remove multiboot1_header_end from symbol table

Its visibility is not needed and just pollute symbol table. Suggested-by: Jan Beulich Signed-off-by: Daniel Kiper --- xen/arch/x86/boot/head.S |2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/xen/arch/x86/boot/head.S

[Xen-devel] [xen-4.7-testing test] 99961: regressions - FAIL

flight 99961 xen-4.7-testing real [real] http://logs.test-lab.xenproject.org/osstest/logs/99961/ Regressions :-( Tests which did not succeed and are blocking, including tests which could not be run: test-armhf-armhf-xl-arndale 7 host-ping-check-xen fail REGR. vs. 99754 Regressions

Re: [Xen-devel] [PATCH v2 1/3] livepach: Add .livepatch.hooks functions and test-case

On Fri, Aug 05, 2016 at 09:35:49AM -0600, Jan Beulich wrote: > >>> On 04.08.16 at 17:49, wrote: > > In general, the hooks provide flexibility when having to deal with > > unforeseen cases, but their application should be rarely required (< > > 10%)." > > But the greater

Re: [Xen-devel] [PATCH v2 3/3] livepatch: Sync cache of build-id before using it first time.

On Fri, Aug 05, 2016 at 09:40:43AM -0600, Jan Beulich wrote: > >>> On 04.08.16 at 17:49, wrote: > > --- a/xen/include/xen/livepatch.h > > +++ b/xen/include/xen/livepatch.h > > @@ -44,7 +44,7 @@ unsigned long livepatch_symbols_lookup_by_name(const char > > *symname); > >

[Xen-devel] [xen-unstable-smoke test] 99970: tolerable all pass - PUSHED

flight 99970 xen-unstable-smoke real [real] http://logs.test-lab.xenproject.org/osstest/logs/99970/ Failures :-/ but no regressions. Tests which did not succeed, but are not blocking: test-amd64-amd64-libvirt 12 migrate-support-checkfail never pass test-armhf-armhf-xl 12

Re: [Xen-devel] monitor access to pages with a specific p2m_type_t

On Fri, Aug 5, 2016 at 10:45 PM, Tamas K Lengyel wrote: > > > On Fri, Aug 5, 2016 at 5:35 AM, sepanta s wrote: > >> >> >> On Tue, Aug 2, 2016 at 8:23 PM, Tamas K Lengyel >> wrote: >> >>> >>> >>> On Tue, Aug 2, 2016 at 12:19 AM,

Re: [Xen-devel] monitor access to pages with a specific p2m_type_t

On Fri, Aug 5, 2016 at 5:35 AM, sepanta s wrote: > > > On Tue, Aug 2, 2016 at 8:23 PM, Tamas K Lengyel > wrote: > >> >> >> On Tue, Aug 2, 2016 at 12:19 AM, sepanta s wrote: >> >>> >>> >>> On Sat, Jul 23, 2016 at 3:49 PM, sepanta

[Xen-devel] [PATCH v2 3/6] libxl: factor out libxl__console_tty_path

No other user yet. Signed-off-by: Wei Liu Acked-by: Ian Jackson --- tools/libxl/libxl.c | 57 ++--- 1 file changed, 37 insertions(+), 20 deletions(-) diff --git a/tools/libxl/libxl.c

[Xen-devel] [PATCH v2 1/6] tools/console: fix help string in client

There is no short '-t' option. Signed-off-by: Wei Liu Acked-by: Ian Jackson --- tools/console/client/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/console/client/main.c b/tools/console/client/main.c index

[Xen-devel] [PATCH v2 0/6] Fix console synchronisation issues

Version 2 of this series. Now it properly handles read and write errors. I also dropped two patches to not move the wait bodge from xenconsole to libxl. Wei Liu (6): tools/console: fix help string in client tools/console: introduce --start-notify-fd option for console client libxl: factor

[Xen-devel] [PATCH v2 6/6] xl: use xenconsole startup protocol

If user asks xl to automatically connect to console when creating a guest, use the new startup protocol before trying to unpause domain so that we don't lose any console output. Signed-off-by: Wei Liu --- v2: properly handle read(2) errors --- tools/libxl/xl_cmdimpl.c | 43

[Xen-devel] [PATCH v2 2/6] tools/console: introduce --start-notify-fd option for console client

The console client will write 0x00 to that fd before entering console loop to indicate its readiness. Signed-off-by: Wei Liu --- v2: properly handle write(2) errors --- tools/console/client/main.c | 22 ++ 1 file changed, 22 insertions(+) diff --git

[Xen-devel] [PATCH v2 4/6] libxl: libxl_{primary_, }console_exec now take notify_fd argument

The new argument will be passed down to xenconsole process, which then uses it to notify readiness. Signed-off-by: Wei Liu --- tools/libxl/libxl.c | 15 +++ tools/libxl/libxl.h | 43 --- 2 files changed, 51 insertions(+),

[Xen-devel] [PATCH v2 5/6] docs: document xenconsole startup protocol

Signed-off-by: Wei Liu Acked-by: Ian Jackson --- docs/misc/console.txt | 6 ++ 1 file changed, 6 insertions(+) diff --git a/docs/misc/console.txt b/docs/misc/console.txt index ed7b795..16da805 100644 --- a/docs/misc/console.txt +++

[Xen-devel] [PATCH v2 04/18] mini-os: add description of x86 memory usage

Add a brief description how the physical and virtual address usage looks like on x86 to include/x86/arch_mm.h Signed-off-by: Juergen Gross Reviewed-by: Wei Liu --- include/x86/arch_mm.h | 20 1 file changed, 20 insertions(+) diff

[Xen-devel] [PATCH v2 18/18] mini-os: balloon up in case of oom

If a memory shortage is detected balloon up. Be careful to always leave some pages free as ballooning up might need some memory, too: - new p2m frames - page tables for addressing new p2m frame - new frame for page allocation bitmap - page table for addressing new page allocation bitmap frame -

[Xen-devel] [PATCH v2 15/18] mini-os: remap p2m list in case of ballooning

In case of enabled ballooning we must be prepared for a growing p2m list. If the maximum memory size of the domain can't be covered by the actual p2m list remap it to the kernel virtual mapping area and leave enough space at the end. Signed-off-by: Juergen Gross ---

Re: [Xen-devel] [PATCH] acpi: Re-license ACPI builder files from GPLv2 to LGPLv2.1

Re: [Xen-devel] [PATCH] acpi: Re-license ACPI builder files from.eml Subject: Re: [Xen-devel] [PATCH] acpi: Re-license ACPI builder files from GPLv2 to LGPLv2.1 From: "Tian, Kevin" > Date: Thu, 4 Aug 2016 06:52:34 + To: Boris Ostrovsky

[Xen-devel] [PATCH v2 07/18] mini-os: add ballooning config item

Add CONFIG_BALLOON defaulting to 'n' as a config item to Mini-OS. Add balloon.c, balloon.h and arch/*/balloon.c for future use. Signed-off-by: Juergen Gross --- V2: Added dummy sources and header --- Makefile | 3 +++ arch/arm/balloon.c | 28

[Xen-devel] [PATCH v2 03/18] mini-os: remove MM_DEBUG code

mm.c contains unused code inside #ifdef MM_DEBUG areas. Its usability is rather questionable and some parts are even wrong (e.g. print_chunks() called with nr_pages > 1000 will clobber an arbitrary stack content with a 0 byte). Remove this code. Signed-off-by: Juergen Gross

[Xen-devel] [PATCH v2 08/18] mini-os: get maximum memory size from hypervisor

Add support for obtaining the maximum memory size from the hypervisor. This will make it possible to support ballooning. Signed-off-by: Juergen Gross --- V2: Moved new stuff to balloon.c --- balloon.c | 22 ++ include/balloon.h | 6 ++ mm.c

[Xen-devel] [PATCH v2 16/18] mini-os: map page allocator's bitmap to virtual kernel area for ballooning

In case of CONFIG_BALLOON the page allocator's bitmap needs some space to be able to grow. Remap it to kernel virtual area if the preallocated area isn't large enough. Signed-off-by: Juergen Gross --- balloon.c | 17 + include/balloon.h | 2 ++

[Xen-devel] [PATCH v2 13/18] mini-os: add function to map one frame

Add a function to map one physical frame to a specified virtual address as read/write. This will be used later multiple times. Signed-off-by: Juergen Gross --- include/arm/arch_mm.h | 2 ++ include/mm.h | 1 + mm.c | 5 + 3 files changed, 8

[Xen-devel] [PATCH v2 10/18] mini-os: remove unused mem_test() function

mem_test() isn't used anywhere and its value is rather questionable with mini-os being in a mature state. Remove the function. Signed-off-by: Juergen Gross --- arch/x86/mm.c | 55 --- 1 file changed, 55 deletions(-) diff

[Xen-devel] [PATCH v2 11/18] mini-os: add checks for out of memory

There are several core functions in Mini-OS not checking for failed memory allocations. Add such checks. Add do_map_frames() dummy function to arm architecture as it will be needed in future for compilations to succeed. Signed-off-by: Juergen Gross --- arch/arm/mm.c | 8

[Xen-devel] [PATCH v2 12/18] mini-os: don't allocate new pages for level 1 p2m tree

When constructing the 3 level p2m tree there is no need to allocate new pages for the level 1 containing the p2m info for all pages. The pages from the linear p2m list constructed by the hypervisor can be used for that purpose. Signed-off-by: Juergen Gross --- arch/x86/mm.c |

[Xen-devel] [PATCH v2 17/18] mini-os: add support for ballooning up

Add support for ballooning the domain up by a specified amount of pages. Following steps are performed: - extending the p2m map - extending the page allocator's bitmap - getting new memory pages from the hypervisor - adding the memory at the current end of guest memory Signed-off-by: Juergen

[Xen-devel] [PATCH v2 00/18] mini-os: support of auto-ballooning

Support ballooning Mini-OS automatically up in case of memory shortage. Do some cleanups, a small correction and add some basic features to lay groundwork for support of ballooning in Mini-OS (patches 1-14). The main visible change is the virtual memory layout: to be able to add memory to the

[Xen-devel] [PATCH v2 06/18] mini-os: let memory allocation fail if no free page available

Instead of panicing when no page can be allocated try to fail the memory allocation by returning NULL instead. Signed-off-by: Juergen Gross Reviewed-by: Wei Liu --- V2: fixed minor style issue --- mm.c | 7 +++ 1 file changed, 7 insertions(+) diff

[Xen-devel] [PATCH v2 09/18] mini-os: modify virtual memory layout for support of ballooning

In order to be able to support ballooning the virtual memory layout of Mini-OS has to be modified: instead of a (nearly) consecutive area used for physical memory mapping, on demand mappings, and heap we need enough spare place for adding new memory. So instead of dynamically place the different

[Xen-devel] [PATCH v2 02/18] mini-os: remove unused alloc_contig_pages() function

alloc_contig_pages() is never used anywhere in mini-os. Remove it. Signed-off-by: Juergen Gross Reviewed-by: Wei Liu --- arch/x86/mm.c | 142 -- include/mm.h | 1 - 2 files changed, 143

[Xen-devel] [PATCH v2 14/18] mini-os: move p2m related macros to header file

In order to be able to use p2m related macros for ballooning move their definitions to arch/x86/mm.h. There is no need to define different macros regarding index masks and number of entries for the different levels, as all levels share the same entry format (a plain mfn). So reduce the number of

[Xen-devel] [PATCH v2 01/18] mini-os: correct first free pfn

The first free pfn available for allocation is calculated by adding the number of page table frames to the pfn of the first page table and then the magic number 3 to account for start info page et al. As the start info page, xenstore page and console page are allocated _before_ the page tables

[Xen-devel] [PATCH v2 05/18] mini-os: add nr_free_pages counter

Add a variable holding the number of available memory pages. This will aid auto-ballooning later. Signed-off-by: Juergen Gross Reviewed-by: Wei Liu --- include/mm.h | 1 + mm.c | 6 ++ 2 files changed, 7 insertions(+) diff --git a/include/mm.h

Re: [Xen-devel] [RFC PATCH 8/8] tools/console: remove 5s bodge in console client

On Fri, Aug 05, 2016 at 05:36:08PM +0100, Wei Liu wrote: > On Fri, Aug 05, 2016 at 05:32:30PM +0100, Ian Jackson wrote: > > Wei Liu writes ("Re: [RFC PATCH 8/8] tools/console: remove 5s bodge in > > console client"): > > > On Fri, Aug 05, 2016 at 05:18:47PM +0100, Ian Jackson wrote: > > > > Well,

Re: [Xen-devel] [PATCH v2] domctl: relax getdomaininfo permissions

On 05/08/16 14:54, Jan Beulich wrote: On 05.08.16 at 15:10, wrote: >> On 05/08/16 12:20, Jan Beulich wrote: >>> I wonder what good the duplication of the returned domain ID does: I'm >>> tempted to remove the one in the command-specific structure. Does >>> anyone

Re: [Xen-devel] [PATCH] gnttab: Add gntdev device mappings for FreeBSD

On Thu, Aug 04, 2016 at 06:23:51PM +0530, Akshay Jaggi wrote: > Add grant table userspace device mappings for > FreeBSD (enables support for qdisk backend > on FreeBSD Dom0). > > Signed-off-by: Akshay Jaggi > --- > tools/include/xen-sys/FreeBSD/gntdev.h | 118

Re: [Xen-devel] [RFC PATCH 8/8] tools/console: remove 5s bodge in console client

On Fri, Aug 05, 2016 at 05:32:30PM +0100, Ian Jackson wrote: > Wei Liu writes ("Re: [RFC PATCH 8/8] tools/console: remove 5s bodge in > console client"): > > On Fri, Aug 05, 2016 at 05:18:47PM +0100, Ian Jackson wrote: > > > Well, except that xenconsole is also on the path. I (now) don't > > >

Re: [Xen-devel] [RFC PATCH 8/8] tools/console: remove 5s bodge in console client

Wei Liu writes ("Re: [RFC PATCH 8/8] tools/console: remove 5s bodge in console client"): > On Fri, Aug 05, 2016 at 05:18:47PM +0100, Ian Jackson wrote: > > Well, except that xenconsole is also on the path. I (now) don't > > understand why this functionality needs to be moved from xenconsole to >

Re: [Xen-devel] [RFC PATCH 8/8] tools/console: remove 5s bodge in console client

On Fri, Aug 05, 2016 at 05:18:47PM +0100, Ian Jackson wrote: > Wei Liu writes ("Re: [RFC PATCH 8/8] tools/console: remove 5s bodge in > console client"): > > On Fri, Aug 05, 2016 at 04:57:33PM +0100, Ian Jackson wrote: > > > In particular, what about > > >xl create /etc/xen/foo.cfg > > >

Re: [Xen-devel] Device model operation hypercall (DMOP, re qemu depriv)

Jan Beulich writes ("Re: Device model operation hypercall (DMOP, re qemu depriv)"): > On 04.08.16 at 13:21, wrote: > > What we cannot do is audit every HVMCTL, fix the class 2 problems, and > > then declare HVMCTL to have the relevant security property, and > >

Re: [Xen-devel] [RFC PATCH 8/8] tools/console: remove 5s bodge in console client

Wei Liu writes ("Re: [RFC PATCH 8/8] tools/console: remove 5s bodge in console client"): > On Fri, Aug 05, 2016 at 04:57:33PM +0100, Ian Jackson wrote: > > In particular, what about > >xl create /etc/xen/foo.cfg > >xl console foo > > ? > > > > I think in this case xenconsole still needs

Re: [Xen-devel] [PATCH v2] libxl: return any serial tty path in libxl_console_get_tty

Wei Liu writes ("Re: [PATCH v2] libxl: return any serial tty path in libxl_console_get_tty"): > On Thu, Aug 04, 2016 at 09:07:45AM +0100, Wei Liu wrote: > > Ian, I think this is a backport candidate. Queued, thanks. Ian. ___ Xen-devel mailing list

Re: [Xen-devel] [RFC PATCH 8/8] tools/console: remove 5s bodge in console client

On Fri, Aug 05, 2016 at 04:57:33PM +0100, Ian Jackson wrote: > Wei Liu writes ("[RFC PATCH 8/8] tools/console: remove 5s bodge in console > client"): > > The bug described in the comment has been fixed in libxl. Since console > > client is a private binary and libxl is the only supported

Re: [Xen-devel] [RFC PATCH 8/8] tools/console: remove 5s bodge in console client

Wei Liu writes ("[RFC PATCH 8/8] tools/console: remove 5s bodge in console client"): > The bug described in the comment has been fixed in libxl. Since console > client is a private binary and libxl is the only supported toolstack in > upstream, remove the bodge to simplify code. I don't think

Re: [Xen-devel] [PATCH v2 2/3] symbols: Generate an xen-sym.map

>>> On 04.08.16 at 17:49, wrote: > You could construct _most_ of the names of the functions > by doing 'nm --defined' but unfortunatly you do not get the > prefix that is added on in Xen . For example: > > $ cat xen-syms.symbols |grep do_domain_pause > 0x82d080104920

Re: [Xen-devel] [RFC PATCH 7/8] xl: use xenconsole startup protocol

Wei Liu writes ("[RFC PATCH 7/8] xl: use xenconsole startup protocol"): > If user asks xl to automatically connect to console when creating a > guest, use the new startup protocol before trying to unpause domain so > that we don't lose any console output. Most of the logic here LGTM. > @@

Re: [Xen-devel] [RFC PATCH 6/8] docs: document xenconsole startup protocol

Wei Liu writes ("[RFC PATCH 6/8] docs: document xenconsole startup protocol"): > Signed-off-by: Wei Liu Oh! Forget my comment on patch 02. about the commit message and/or comment. Maybe you want to squash this one into there, but either way: > +The xenconsole program

Re: [Xen-devel] [RFC PATCH 5/8] libxl: libxl_{primary_, }console_exec now take notify_fd argument

Ian Jackson writes ("Re: [RFC PATCH 5/8] libxl: libxl_{primary_,}console_exec now take notify_fd argument"): > Wei Liu writes ("[RFC PATCH 5/8] libxl: libxl_{primary_,}console_exec now > take notify_fd argument"): > > The new argument will be passed down to xenconsole process, which then > >

Re: [Xen-devel] [RFC PATCH 5/8] libxl: libxl_{primary_, }console_exec now take notify_fd argument

Wei Liu writes ("[RFC PATCH 5/8] libxl: libxl_{primary_,}console_exec now take notify_fd argument"): > The new argument will be passed down to xenconsole process, which then > uses it to notify readiness. ... > int libxl_console_exec(libxl_ctx *ctx, uint32_t domid, int cons_num, > -

Re: [Xen-devel] [RFC PATCH 4/8] libxl: wait up to 5s in libxl_console_exec for xenconsoled

Wei Liu writes ("[RFC PATCH 4/8] libxl: wait up to 5s in libxl_console_exec for xenconsoled"): > Wait until the tty node is available before exec'ing xenconsole. You shouldn't poll. We have a perfectly good xenstore watch mechanism. You could either: (a) do something ad-hoc with poll(), the

Re: [Xen-devel] [RFC PATCH 1/8] tools/console: fix help string in client

Wei Liu writes ("[RFC PATCH 1/8] tools/console: fix help string in client"): > There is no short '-t' option. Acked-by: Ian Jackson ___ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel

Re: [Xen-devel] [RFC PATCH 3/8] libxl: factor out libxl__console_tty_path

Wei Liu writes ("[RFC PATCH 3/8] libxl: factor out libxl__console_tty_path"): > No user yet. You mean no _other_ user! With that change to the commit message, Acked-by: Ian Jackson Ian. ___ Xen-devel mailing list

Re: [Xen-devel] [RFC PATCH 2/8] tools/console: introduce --start-notify-fd option for console client

Wei Liu writes ("[RFC PATCH 2/8] tools/console: introduce --start-notify-fd option for console client"): > The console client will write 0x00 to that fd before entering console > loop to indicate its readiness. This could sensibly be in a comment in the code. > + if (start_notify_fd != -1)

Re: [Xen-devel] [PATCH v2 3/3] livepatch: Sync cache of build-id before using it first time.

>>> On 04.08.16 at 17:49, wrote: > --- a/xen/include/xen/livepatch.h > +++ b/xen/include/xen/livepatch.h > @@ -44,7 +44,7 @@ unsigned long livepatch_symbols_lookup_by_name(const char > *symname); > bool_t is_patch(const void *addr); > int xen_build_id_check(const

Re: [Xen-devel] [PATCH v2 1/3] livepach: Add .livepatch.hooks functions and test-case

>>> On 04.08.16 at 17:49, wrote: > In general, the hooks provide flexibility when having to deal with > unforeseen cases, but their application should be rarely required (< > 10%)." But the greater flexibility of course comes with increased chances of screwing things up.

Re: [Xen-devel] Xen 4.6.1 crash with altp2m enabled bydefault

>>> On 05.08.16 at 14:51, wrote: > According to the xen dmesg > > (XEN) RIP:e008:[] vmx_vmenter_helper+0x27e/0x30a > (XEN) RFLAGS: 00010003 CONTEXT: hypervisor > (XEN) rax: 8005003b rbx: 8300e72fc000 rcx: > (XEN) rdx:

Re: [Xen-devel] [PATCH 2/3] xen: Have schedulers revise initial placement

>>> On 05.08.16 at 16:09, wrote: > On Fri, 2016-08-05 at 07:24 -0600, Jan Beulich wrote: >> > > > On 01.08.16 at 14:32, wrote: >> > On Mon, 2016-08-01 at 04:40 -0600, Jan Beulich wrote: >> > > > > > On 15.07.16 at 20:02,

Re: [Xen-devel] live migration to qemu.git fails

On Fri, Aug 05, Anthony PERARD wrote: > On Fri, Aug 05, 2016 at 03:37:32PM +0200, Olaf Hering wrote: > If you compile older qemu out if xen tree, you would need to at least > add: > --extra-cflags="-DXC_WANT_COMPAT_EVTCHN_API=1 -DXC_WANT_COMPAT_GNTTAB_API=1 > -DXC_WANT_COMPAT_MAP_FOREIGN_API=1"

Re: [Xen-devel] [PATCH 2/3] xen: Have schedulers revise initial placement

On Fri, 2016-08-05 at 07:24 -0600, Jan Beulich wrote: > > > > On 01.08.16 at 14:32, wrote: > > On Mon, 2016-08-01 at 04:40 -0600, Jan Beulich wrote: > > > > > > On 15.07.16 at 20:02, wrote: > > > > Signed-off-by: George Dunlap

Re: [Xen-devel] [PATCH 3/3] x86/microcode: Avoid undefined behaviour from signed integer overflow

>>> On 05.08.16 at 15:50, wrote: > --- a/xen/arch/x86/microcode_intel.c > +++ b/xen/arch/x86/microcode_intel.c > @@ -143,7 +143,8 @@ static int microcode_sanity_check(void *mc) > struct extended_sigtable *ext_header = NULL; > struct extended_signature

Re: [Xen-devel] [PATCH 1/3] xen/common: Avoid undefined behaviour by shifting into a sign bit

On 05/08/16 15:04, Jan Beulich wrote: On 05.08.16 at 15:50, wrote: >> --- a/xen/common/domctl.c >> +++ b/xen/common/domctl.c >> @@ -188,7 +188,7 @@ void getdomaininfo(struct domain *d, struct >> xen_domctl_getdomaininfo *info) >>

Re: [Xen-devel] [PATCH 2/3] xen/x86: Avoid undefined behaviour by shifting into a sign bit

>>> On 05.08.16 at 15:50, wrote: > --- a/xen/arch/x86/cpu/common.c > +++ b/xen/arch/x86/cpu/common.c > @@ -476,7 +476,7 @@ void detect_extended_topology(struct cpuinfo_x86 *c) > sub_index++; > } while ( LEAFB_SUBTYPE(ecx) != INVALID_TYPE ); > > -

Re: [Xen-devel] [PATCH 1/3] xen/common: Avoid undefined behaviour by shifting into a sign bit

>>> On 05.08.16 at 15:50, wrote: > --- a/xen/common/domctl.c > +++ b/xen/common/domctl.c > @@ -188,7 +188,7 @@ void getdomaininfo(struct domain *d, struct > xen_domctl_getdomaininfo *info) > (d->controller_pause_count > 0 ? XEN_DOMINF_paused: 0) | >

Re: [Xen-devel] live migration to qemu.git fails

On Fri, Aug 05, 2016 at 03:37:32PM +0200, Olaf Hering wrote: > On Wed, Aug 03, Olaf Hering wrote: > > > On Wed, Aug 03, Anthony PERARD wrote: > > > > > Haven't you try to create a guest with Xen 4.5 and qemu-xen-4.5, and > > > then migrate to Xen 4.7 with QEMU-2.6/master? > > > > In the end I

Re: [Xen-devel] [PATCH] CODING_STYLE: Allow single-sentence comments without full stops

On Thu, Aug 4, 2016 at 11:25 AM, Ian Jackson wrote: > George Dunlap writes ("[PATCH] CODING_STYLE: Allow single-sentence comments > without full stops"): >> One of the common ways in which contributors trip up over the >> CODING_STYLE guides is by not putting a full

Re: [Xen-devel] [PATCH v2] domctl: relax getdomaininfo permissions

>>> On 05.08.16 at 15:10, wrote: > On 05/08/16 12:20, Jan Beulich wrote: >> I wonder what good the duplication of the returned domain ID does: I'm >> tempted to remove the one in the command-specific structure. Does >> anyone have insight into why it was done that way?

[Xen-devel] [PATCH 3/3] x86/microcode: Avoid undefined behaviour from signed integer overflow

The checksum should be calculated using unsigned 32bit integers, as it is intended to overflow and end at 0. Signed-off-by: Andrew Cooper --- CC: Jan Beulich CC: Kevin Tian CC: Jun Nakajima ---

[Xen-devel] [PATCH 2/3] xen/x86: Avoid undefined behaviour by shifting into a sign bit

Signed-off-by: Andrew Cooper --- CC: Jan Beulich --- xen/arch/x86/apic.c | 2 +- xen/arch/x86/cpu/common.c | 2 +- xen/arch/x86/x86_64/traps.c | 2 +- xen/include/asm-x86/apicdef.h | 2 +- 4 files changed, 4 insertions(+), 4

[Xen-devel] [PATCH 1/3] xen/common: Avoid undefined behaviour by shifting into a sign bit

Signed-off-by: Andrew Cooper --- CC: Jan Beulich CC: George Dunlap CC: Konrad Rzeszutek Wilk CC: Stefano Stabellini CC: Tim Deegan --- xen/common/domctl.c

[Xen-devel] [qemu-mainline baseline-only test] 66921: tolerable FAIL

This run is configured for baseline tests only. flight 66921 qemu-mainline real [real] http://osstest.xs.citrite.net/~osstest/testlogs/logs/66921/ Failures :-/ but no regressions. Regressions which are regarded as allowable (not blocking): test-amd64-amd64-qemuu-nested-intel 16

Re: [Xen-devel] live migration to qemu.git fails

On Wed, Aug 03, Olaf Hering wrote: > On Wed, Aug 03, Anthony PERARD wrote: > > > Haven't you try to create a guest with Xen 4.5 and qemu-xen-4.5, and > > then migrate to Xen 4.7 with QEMU-2.6/master? > > In the end I tried xen-4.5/6/7/8 as source and their qemu-xen, and > migrated to

Re: [Xen-devel] [RFC Design Doc v2] Add vNVDIMM support for Xen

> > > As above, if linux driver detects the signature "NVDIMM_PFN_INFO" and > > > a matched checksum, it will know it's safe to write to the reserved > > > area. Otherwise, it will treat the pmem namespace as a raw device and > > > store page struct's in the normal RAM. > > > > OK, so my worry is

Re: [Xen-devel] [PATCH 2/3] xen: Have schedulers revise initial placement

>>> On 01.08.16 at 14:32, wrote: > On Mon, 2016-08-01 at 04:40 -0600, Jan Beulich wrote: >> > > > On 15.07.16 at 20:02, wrote: >> > >> > To solve this, when inserting a vcpu, always call the per-scheduler >> > "pick" function to revise the

Re: [Xen-devel] [PATCH v2] domctl: relax getdomaininfo permissions

On 05/08/16 12:20, Jan Beulich wrote: > Qemu needs access to this for the domain it controls, both due to it > being used by xc_domain_memory_mapping() (which qemu calls) and the > explicit use in hw/xenpv/xen_domainbuild.c:xen_domain_poll(). Extend > permissions to that of any "ordinary" domctl:

[Xen-devel] [qemu-mainline test] 99953: regressions - FAIL

flight 99953 qemu-mainline real [real] http://logs.test-lab.xenproject.org/osstest/logs/99953/ Regressions :-( Tests which did not succeed and are blocking, including tests which could not be run: test-amd64-i386-qemuu-rhel6hvm-amd 5 xen-install fail REGR. vs. 99944 Regressions which

Re: [Xen-devel] Xen 4.6.1 crash with altp2m enabled bydefault

According to the xen dmesg (XEN) RIP:e008:[] vmx_vmenter_helper+0x27e/0x30a (XEN) RFLAGS: 00010003 CONTEXT: hypervisor (XEN) rax: 8005003b rbx: 8300e72fc000 rcx: (XEN) rdx: 6c00 rsi: 830617fd7fc0 rdi: 8300e6fc (XEN) rbp:

  1   2   >