>>> On 27.09.16 at 16:03, wrote:
> On 15/09/16 07:43, Jan Beulich wrote:
> On 14.09.16 at 19:30, wrote:
+#define X86EMUL_OPC_PFX_MASK 0x0300
+# define X86EMUL_OPC_66(ext, byte) (X86EMUL_OPC(ext, byte) | 0x0100)
+# define X86EMUL_OPC_F3(ext, byte) (X86EMUL_OP
>>> On 27.09.16 at 19:43, wrote:
> Finally found the vmfunc opcode page in Vol 3 30.3, VMX Instruction Reference.
> Agreed, there's no mention of prefixes, "pfx", on this page. It appears
> that the other VMX instructions in this section don't mention prefixes either.
> Looking at Table A-6 "Opco
..., complete the decoder, leverage decoding for SVM instruction
sizing and PV 32-bit call gate emulation, and use the emulator for
PV priv-op handling.
01: x86emul: split instruction decoding from execution
02: x86emul: fetch all insn bytes during the decode phase
03: x86emul: track only rIP in e
This is only the mechanical part, a subsequent patch will make non-
mechanical adjustments to actually do all decoding in this new
function.
Signed-off-by: Jan Beulich
---
v2: Fix a coding style issue.
--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -
This way we can offer to callers the service of just sizing
instructions, and we also can better guarantee not to raise the wrong
fault due to not having read all relevant bytes.
Signed-off-by: Jan Beulich
---
v2: Rename x86_decode_base() -> x86_decode_onebyte().
--- a/xen/arch/x86/x86_emulate/x
Now that all decoding happens in x86_decode() there's no need to keep
the local registers copy in struct x86_emulate_state. Only rIP gets
updated in the decode phase, so only that register needs tracking
there. All other (read-only) registers can be read from the original
structure (but sadly, due
This way we can at least size (and e.g. skip) them if needed, and we
also won't raise the wrong fault due to not having read all relevant
bytes.
Signed-off-by: Jan Beulich
---
v2: Add a comment.
--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -279,6 +
This way we can at least size (and e.g. skip) them if needed, and we
also won't raise the wrong fault due to not having read all relevant
bytes.
This at once adds correct raising of #UD for the three "ud" flavors
(Intel names only "ud2", but AMD names all three of them in their
opcode maps), as th
This way we can at least size (and e.g. skip) them if needed, and we
also won't raise the wrong fault due to not having read all relevant
bytes.
Signed-off-by: Jan Beulich
---
TBD: I'm kind of undecided whether to right away propagate evex.R into
modrm_reg (and then also deal with the new me
This representation is then being made available to interested callers,
to facilitate replacing their custom decoding.
This entails combining the three main switch statements into one.
Signed-off-by: Jan Beulich
---
v2: Extend comments. Use uint8_t cast in X86EMUL_OPC(). Rename
X86EMUL_OPC_K
... instead of custom handling. To facilitate this break out init code
from _hvm_emulate_one() into the new hvm_emulate_init(), and make
hvmemul_insn_fetch( globally available.
Signed-off-by: Jan Beulich
---
v2: Add comment to caller field. Rename REG_POISON to PTR_POISON. Align
opc_tab[] ini
... instead of custom handling. Note that we can't use generic
emulation, as the emulator's far branch support is rather rudimentary
at this point in time.
Signed-off-by: Jan Beulich
--- a/xen/arch/x86/traps.c
+++ b/xen/arch/x86/traps.c
@@ -28,6 +28,7 @@
#include
#include
#include
+#includ
This is in preparation for using the generic emulator here.
Some care is needed temporarily to not unduly alter guest register
state: The local variable "res" can only go away once this code got
fully switched over to using x86_emulate().
Also switch to IS_ERR_VALUE() instead of (incorrectly) ope
This is in preparation for using the generic emulator here.
Signed-off-by: Jan Beulich
---
v2: Re-base.
--- a/xen/arch/x86/traps.c
+++ b/xen/arch/x86/traps.c
@@ -2386,6 +2386,345 @@ static inline uint64_t guest_misc_enable
return val;
}
+static inline bool is_cpufreq_controller(const str
This is in preparation for using the generic emulator here.
Signed-off-by: Jan Beulich
--- a/xen/arch/x86/traps.c
+++ b/xen/arch/x86/traps.c
@@ -2255,6 +2255,107 @@ unsigned long guest_to_host_gpr_switch(u
void (*pv_post_outb_hook)(unsigned int port, u8 value);
+static int priv_op_read_cr(u
This is a prereq for switching PV privileged op emulation to the
generic instruction emulator. Since handle_xsetbv() is already capable
of dealing with all guest kinds, avoid introducing another hook here.
Signed-off-by: Jan Beulich
---
v2: Explicitly generate #UD when vex.pfx is non-zero.
--- a
There's a new emulator return code being added to allow bypassing
certain operations (see the code comment). Its handling in the epilogue
code involves moving the raising of the single step trap until after
registers were updated. This should probably have been that way from
the beginning, to allow
Sort the special case opcode 0f01 entries numerically, insert blank
lines between each of the cases, and properly place opening braces.
No functional change.
Signed-off-by: Jan Beulich
--- a/xen/arch/x86/x86_emulate/x86_emulate.c
+++ b/xen/arch/x86/x86_emulate/x86_emulate.c
@@ -4195,6 +4195,14
Especially for x86_insn_operand_ea() to return dependable segment
information even when the caller didn't consider applicability, we
shouldn't have ea.type start out as OP_MEM. Make it OP_NONE instead,
and set it to OP_MEM when we actually encounter memory like operands.
This requires to eliminate
On 27/09/16 17:57, Roger Pau Monne wrote:
> Introduce a new %pB format specifier to print sizes (in bytes) in a
Code suggests it is pZ.
> human-readable form.
>
> Signed-off-by: Roger Pau Monné
> ---
> Cc: Andrew Cooper
> Cc: George Dunlap
> Cc: Ian Jackson
> Cc: Jan Beulich
> Cc: Konrad Rz
grub_printf() is supporting only a very limited number of formats.
Especially some error messages suffer from that, e.g. %lx won't work.
Switch to use printk() for error messages instead.
Signed-off-by: Juergen Gross
---
stubdom/grub/kexec.c | 22 +++---
1 file changed, 11 insert
>>> On 28.09.16 at 09:59, wrote:
> 01: x86emul: split instruction decoding from execution
> 02: x86emul: fetch all insn bytes during the decode phase
> 03: x86emul: track only rIP in emulator state
> 04: x86emul: complete decoding of two-byte instructions
> 05: x86emul: add XOP decoding
> 06: x86e
>>> On 27.09.16 at 19:49, wrote:
> On Mon, Sep 26, 2016 at 07:37:45AM -0600, Jan Beulich wrote:
>> >>> On 23.09.16 at 23:47, wrote:
>> > --- a/xen/common/efi/boot.c
>> > +++ b/xen/common/efi/boot.c
>> > @@ -79,6 +79,10 @@ static size_t wstrlen(const CHAR16 * s);
>> > static int set_color(u32 mas
>>> On 27.09.16 at 20:11, wrote:
> On Mon, Sep 26, 2016 at 07:47:42AM -0600, Jan Beulich wrote:
>> >>> On 23.09.16 at 23:47, wrote:
>> > This way Xen can be loaded on EFI platforms using GRUB2 and
>> > other boot loaders which support multiboot2 protocol.
>> >
>> > Signed-off-by: Daniel Kiper
>>
>>> On 27.09.16 at 20:21, wrote:
> On Mon, Sep 26, 2016 at 09:12:40AM -0600, Jan Beulich wrote:
>> >>> On 26.09.16 at 16:40, wrote:
>> > On 26/09/16 15:33, Jan Beulich wrote:
>> > On 26.09.16 at 16:19, wrote:
>> >>> On 23/09/16 22:47, Daniel Kiper wrote:
>> +/*
>> +
>>> On 27.09.16 at 21:55, wrote:
> On Mon, Sep 26, 2016 at 09:03:30AM -0600, Jan Beulich wrote:
>> >>> On 23.09.16 at 23:47, wrote:
>> > @@ -426,32 +453,65 @@ trampoline_bios_setup:
>> > xor %cl, %cl
>> >
>> > trampoline_setup:
>> > +/*
>> > + * Called on legacy BIOS
flight 101175 xen-unstable real [real]
http://logs.test-lab.xenproject.org/osstest/logs/101175/
Failures :-/ but no regressions.
Tests which are failing intermittently (not blocking):
test-armhf-armhf-xl-rtds 15 guest-start/debian.repeat fail pass in 101169
test-armhf-armhf-xl-arndale 15
>>> On 27.09.16 at 16:43, wrote:
> If the guest is booted with 'pci' we nicely expand the MMIO region below
> 4GB and try to fit in the BARs in there. If that fails (not enough
> space) we move it above the memory (64-bit). And throughout all of this
> we also update the _CRS field to cover these
flight 101176 libvirt real [real]
http://logs.test-lab.xenproject.org/osstest/logs/101176/
Failures :-/ but no regressions.
Tests which did not succeed, but are not blocking:
test-armhf-armhf-libvirt-xsm 12 migrate-support-checkfail never pass
test-armhf-armhf-libvirt-xsm 14 guest-sav
At 17:56 +0200 on 27 Sep (1474999018), Roger Pau Monne wrote:
> Return should be an int, and the number of pages should be an unsigned long.
>
> Signed-off-by: Roger Pau Monné
Both those changes seem fine to me. Since you're changing the return
types to int, can you please also change the two c
>>> On 28.09.16 at 08:48, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Monday, September 26, 2016 8:10 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>> de...@l
On Wed, Sep 28, 2016 at 02:57:10AM -0600, Jan Beulich wrote:
> >>> On 27.09.16 at 20:11, wrote:
> > On Mon, Sep 26, 2016 at 07:47:42AM -0600, Jan Beulich wrote:
> >> >>> On 23.09.16 at 23:47, wrote:
> >> > This way Xen can be loaded on EFI platforms using GRUB2 and
> >> > other boot loaders which
the struct tmem_op.
No functional change. But it makes the code so much easier
to read.
Note: We still have to do this awkward 'guest_handle_cast'
otherwise it will not compile on ARM - which defines _two_
of these macros (__guest_handle_64_xen_sysctl_tmem_client_t
and __guest_handle_xen_sysctl_t
It is not used by anything.
Signed-off-by: Konrad Rzeszutek Wilk
---
Cc: Ian Jackson
Cc: Wei Liu
v1: First submission
---
docs/man/xl.pod.1.in | 4
tools/libxc/xc_tmem.c | 13 +++--
tools/libxl/libxl.c | 4 +---
tools/libxl/xl_cmdtable.c
This config is not defined anywhere but it makes it way
easier to figure out what code to deal with.
Signed-off-by: Konrad Rzeszutek Wilk
---
v1: First submission.
---
xen/common/tmem.c | 57 ++
xen/common/tmem_control.c | 2 ++
xen/common/t
in its own structure. This paves the way to make only
one hypercall to retrieve/set this information instead of multiple
ones.
Signed-off-by: Konrad Rzeszutek Wilk
---
v1: First submission.
---
xen/common/tmem.c | 40 +---
xen/common/tmem_control.c
. which is actually dependent on CONFIG_TMEM_DEDUP
Signed-off-by: Konrad Rzeszutek Wilk
---
v1: First submission.
---
xen/common/tmem.c | 12 ++--
xen/common/tmem_xen.c | 4
xen/include/xen/tmem_xen.h | 4
3 files changed, 18 insertions(+), 2 deletions(-)
diff
That is what they are used for. Lets make it more clear.
Of all the various sub-commands, the only one that needed
semantic change is XEN_SYSCTL_TMEM_OP_SAVE_BEGIN. That in the
past used 'arg1', and now we are moving it to use 'arg'.
Since that code is only used during migration which is tied
to t
Hey!
This batch of fixes slowly marches toward ripping out pieces
of code in tmem that are hard to maintain and improve on code
that was orginacally developed.
I had hoped that I would have had the migration support all
working, but it took longer than I thought to get to this
point (and migratio
Couple of reasons:
- It can lead to security issues (see row-hammer, KSM and such
attacks).
- Code is quite complex.
- Deduplication is good if the pages themselves are the same
but that is hardly guaranteed.
- We got some gains (if pages are deduped) but at the cost of
making code les
is called. If they are different from what the hypervisor
can support we will get the appropiate errors.
Signed-off-by: Konrad Rzeszutek Wilk
---
Cc: Ian Jackson
Cc: Wei Liu
v1: First submission.
---
tools/libxc/xc_tmem.c | 1 -
xen/common/tmem_control.c | 6 ++
xen/include/publi
These operations are used during the save process of migration.
Instead of doing 64 hypercalls lets do just one. We modify
the 'struct tmem_client' structure (used in
XEN_SYSCTL_TMEM_OP_[GET|SET]_CLIENT_INFO) to have an extra field
'nr_pools'. Armed with that the code slurping up pages from the
hyp
return values. For success they used to be 1 ([SAVE,RESTORE]_BEGIN),
0 if guest did not have any tmem (but only for SAVE_BEGIN), and
-1 for any type of failure.
And SAVE_END (which you would think would mirror SAVE_BEGIN)
had 0 for success and -1 if guest did not any tmem enabled for it.
This is
The only thing this hypercall returns is TMEM_SPEC_VERSION.
The comment around is also misleading - this call does not
do any domain operation.
Signed-off-by: Konrad Rzeszutek Wilk
---
Cc: Ian Jackson
Cc: Wei Liu
v1: Initial submission.
---
tools/libxc/xc_tmem.c | 4
1 file changed, 4
Specifically:
XEN_SYSCTL_TMEM_OP_SET_[WEIGHT,COMPRESS] are now done via:
XEN_SYSCTL_TMEM_SET_CLIENT_INFO
and XEN_SYSCTL_TMEM_OP_SAVE_GET_[VERSION,MAXPOOLS,
CLIENT_WEIGHT, CLIENT_FLAGS] can now be retrieved via:
XEN_SYSCTL_TMEM_GET_CLIENT_INFO
All this information is now in 'struct tmem_clien
Commit e35295d724d64e6e025ec31c2e510e448b8641d1 ("mini-os: fix builds
with uncommon config settings") introduced a bug which will break
the build, as xc_minios.c will have defined those macros via
xc_private.h.
Fix this by defining the macros only if compiling a minios source.
Signed-off-by: Juer
>>> On 28.09.16 at 08:50, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Monday, September 26, 2016 7:47 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>> de...@l
On Wed, Sep 28, 2016 at 03:21:08AM -0600, Jan Beulich wrote:
> >>> On 27.09.16 at 16:43, wrote:
> > If the guest is booted with 'pci' we nicely expand the MMIO region below
> > 4GB and try to fit in the BARs in there. If that fails (not enough
> > space) we move it above the memory (64-bit). And t
On Wed, Sep 28, 2016 at 03:06:31AM -0600, Jan Beulich wrote:
> >>> On 27.09.16 at 21:55, wrote:
> > On Mon, Sep 26, 2016 at 09:03:30AM -0600, Jan Beulich wrote:
> >> >>> On 23.09.16 at 23:47, wrote:
> >> > @@ -426,32 +453,65 @@ trampoline_bios_setup:
> >> > xor %cl, %cl
> >> >
> >> >
>>> On 28.09.16 at 08:51, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Monday, September 26, 2016 8:58 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>> de...@l
On Tue, Sep 27, 2016 at 11:43:38AM -0700, Shannon Zhao wrote:
>
>
> On 2016/9/27 9:35, Wei Liu wrote:
> >On Tue, Sep 27, 2016 at 09:01:00AM -0700, Shannon Zhao wrote:
> >>
> >>
> >>On 2016/9/27 2:41, Wei Liu wrote:
> >>>On Mon, Sep 26, 2016 at 02:54:55PM -0700, Shannon Zhao wrote:
>
>
>
On Wed, Sep 28, 2016 at 11:50:04AM +0200, Juergen Gross wrote:
> Commit e35295d724d64e6e025ec31c2e510e448b8641d1 ("mini-os: fix builds
> with uncommon config settings") introduced a bug which will break
> the build, as xc_minios.c will have defined those macros via
> xc_private.h.
>
> Fix this by
On 28/09/16 12:01, Wei Liu wrote:
> On Wed, Sep 28, 2016 at 11:50:04AM +0200, Juergen Gross wrote:
>> Commit e35295d724d64e6e025ec31c2e510e448b8641d1 ("mini-os: fix builds
>> with uncommon config settings") introduced a bug which will break
>> the build, as xc_minios.c will have defined those macro
On Wed, Sep 28, 2016 at 12:06:02PM +0200, Juergen Gross wrote:
> On 28/09/16 12:01, Wei Liu wrote:
> > On Wed, Sep 28, 2016 at 11:50:04AM +0200, Juergen Gross wrote:
> >> Commit e35295d724d64e6e025ec31c2e510e448b8641d1 ("mini-os: fix builds
> >> with uncommon config settings") introduced a bug whic
On 28/09/16 12:10, Wei Liu wrote:
> On Wed, Sep 28, 2016 at 12:06:02PM +0200, Juergen Gross wrote:
>> On 28/09/16 12:01, Wei Liu wrote:
>>> On Wed, Sep 28, 2016 at 11:50:04AM +0200, Juergen Gross wrote:
Commit e35295d724d64e6e025ec31c2e510e448b8641d1 ("mini-os: fix builds
with uncommon co
>>> On 28.09.16 at 11:39, wrote:
> On Wed, Sep 28, 2016 at 02:57:10AM -0600, Jan Beulich wrote:
>> >>> On 27.09.16 at 20:11, wrote:
>> > On Mon, Sep 26, 2016 at 07:47:42AM -0600, Jan Beulich wrote:
>> >> >>> On 23.09.16 at 23:47, wrote:
>> >> > +/*
>> >> > + * Initialize BSS (no
>>> On 28.09.16 at 11:56, wrote:
> On Wed, Sep 28, 2016 at 03:06:31AM -0600, Jan Beulich wrote:
>> >>> On 27.09.16 at 21:55, wrote:
>> > On Mon, Sep 26, 2016 at 09:03:30AM -0600, Jan Beulich wrote:
>> >> >>> On 23.09.16 at 23:47, wrote:
>> >> > +1: mov %eax,sym_fs(l2_bootmap)-8(%ebx,%ec
On 09/23/2016 04:59 PM, Konrad Rzeszutek Wilk wrote:
On Fri, Sep 23, 2016 at 11:37:27AM -0400, Konrad Rzeszutek Wilk wrote:
On Fri, Sep 23, 2016 at 03:36:27PM +0100, Ross Lagerwall wrote:
On 09/21/2016 06:32 PM, Konrad Rzeszutek Wilk wrote:
If the distance is too big we are in trouble - as our
>>> On 27.09.16 at 16:46, wrote:
> --- a/tools/firmware/hvmloader/acpi/Makefile
> +++ b/tools/firmware/hvmloader/acpi/Makefile
> @@ -17,7 +17,8 @@
> XEN_ROOT = $(CURDIR)/../../../..
> include $(XEN_ROOT)/tools/firmware/Rules.mk
>
> -C_SRC = build.c dsdt_anycpu.c dsdt_15cpu.c static_tables.c
>
flight 101180 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/101180/
Perfect :-)
All tests in this flight passed as required
version targeted for testing:
ovmf ab970515d2c6ec657fceab0ce571054bb43a22f2
baseline version:
ovmf eab26788156436a549610
flight 101181 xen-unstable-coverity real [real]
http://logs.test-lab.xenproject.org/osstest/logs/101181/
Perfect :-)
All tests in this flight passed as required
version targeted for testing:
xen 1e75ed8b64bc1a9b47e540e6f100f17ec6d97f1b
baseline version:
xen a75f
>>> On 27.09.16 at 17:27, wrote:
> --- a/xen/include/public/hvm/params.h
> +++ b/xen/include/public/hvm/params.h
> @@ -30,6 +30,7 @@
> */
>
> #define HVM_PARAM_CALLBACK_IRQ 0
> +#define HVM_PARAM_CALLBACK_IRQ_TYPE_MASK 0xFF00ULL
Well, I have to admit that I'm somewhat disappointe
Juergen Gross, on Wed 28 Sep 2016 10:31:58 +0200, wrote:
> grub_printf() is supporting only a very limited number of formats.
> Especially some error messages suffer from that, e.g. %lx won't work.
> Switch to use printk() for error messages instead.
>
> Signed-off-by: Juergen Gross
Acked-by: Sa
On Wed, Sep 28, 2016 at 10:31:58AM +0200, Juergen Gross wrote:
> grub_printf() is supporting only a very limited number of formats.
> Especially some error messages suffer from that, e.g. %lx won't work.
> Switch to use printk() for error messages instead.
>
> Signed-off-by: Juergen Gross
Acked
On Wed, Sep 28, 2016 at 05:42:22AM -0400, Konrad Rzeszutek Wilk wrote:
> the struct tmem_op.
>
> No functional change. But it makes the code so much easier
> to read.
>
> Note: We still have to do this awkward 'guest_handle_cast'
> otherwise it will not compile on ARM - which defines _two_
> of t
On Wed, Sep 28, 2016 at 05:42:16AM -0400, Konrad Rzeszutek Wilk wrote:
> It is not used by anything.
>
> Signed-off-by: Konrad Rzeszutek Wilk
Acked-by: Wei Liu
> diff --git a/xen/include/public/sysctl.h b/xen/include/public/sysctl.h
> index 8197c14..847ec35 100644
> --- a/xen/include/public/sy
On Wed, Sep 28, 2016 at 05:42:15AM -0400, Konrad Rzeszutek Wilk wrote:
> The only thing this hypercall returns is TMEM_SPEC_VERSION.
>
> The comment around is also misleading - this call does not
> do any domain operation.
>
> Signed-off-by: Konrad Rzeszutek Wilk
Acked-by: Wei Liu
___
On Wed, Sep 28, 2016 at 05:42:23AM -0400, Konrad Rzeszutek Wilk wrote:
> is called. If they are different from what the hypervisor
> can support we will get the appropiate errors.
>
> Signed-off-by: Konrad Rzeszutek Wilk
>
> ---
> Cc: Ian Jackson
> Cc: Wei Liu
>
> v1: First submission.
> ---
On Wed, Sep 28, 2016 at 06:02:44AM +0200, Juergen Gross wrote:
> When trying to boot a kernel with the p2m list not mapped by the
> initial kernel mapping it can happen that pvgrub is failing as it is
> keeping some page tables mapped.
>
> Unmap the additional page tables created for the special p
On Wed, Sep 28, 2016 at 05:42:21AM -0400, Konrad Rzeszutek Wilk wrote:
> Specifically:
>
> XEN_SYSCTL_TMEM_OP_SET_[WEIGHT,COMPRESS] are now done via:
>
> XEN_SYSCTL_TMEM_SET_CLIENT_INFO
>
> and XEN_SYSCTL_TMEM_OP_SAVE_GET_[VERSION,MAXPOOLS,
> CLIENT_WEIGHT, CLIENT_FLAGS] can now be retrieved vi
On Wed, Sep 28, 2016 at 05:42:24AM -0400, Konrad Rzeszutek Wilk wrote:
> return values. For success they used to be 1 ([SAVE,RESTORE]_BEGIN),
> 0 if guest did not have any tmem (but only for SAVE_BEGIN), and
> -1 for any type of failure.
>
> And SAVE_END (which you would think would mirror SAVE_BE
On Wed, Sep 28, 2016 at 05:42:26AM -0400, Konrad Rzeszutek Wilk wrote:
> These operations are used during the save process of migration.
> Instead of doing 64 hypercalls lets do just one. We modify
> the 'struct tmem_client' structure (used in
> XEN_SYSCTL_TMEM_OP_[GET|SET]_CLIENT_INFO) to have an
On Wed, Sep 28, 2016 at 05:42:25AM -0400, Konrad Rzeszutek Wilk wrote:
> That is what they are used for. Lets make it more clear.
>
> Of all the various sub-commands, the only one that needed
> semantic change is XEN_SYSCTL_TMEM_OP_SAVE_BEGIN. That in the
> past used 'arg1', and now we are moving
Juergen Gross, on Wed 28 Sep 2016 11:50:04 +0200, wrote:
> Commit e35295d724d64e6e025ec31c2e510e448b8641d1 ("mini-os: fix builds
> with uncommon config settings") introduced a bug which will break
> the build, as xc_minios.c will have defined those macros via
> xc_private.h.
>
> Fix this by defini
On Wed, Sep 28, 2016 at 01:07:36PM +0200, Samuel Thibault wrote:
> Juergen Gross, on Wed 28 Sep 2016 11:50:04 +0200, wrote:
> > Commit e35295d724d64e6e025ec31c2e510e448b8641d1 ("mini-os: fix builds
> > with uncommon config settings") introduced a bug which will break
> > the build, as xc_minios.c w
flight 101177 qemu-mainline real [real]
http://logs.test-lab.xenproject.org/osstest/logs/101177/
Failures :-/ but no regressions.
Regressions which are regarded as allowable (not blocking):
test-amd64-i386-xl-qemuu-win7-amd64 16 guest-stop fail like 101172
test-amd64-amd64-xl-rtds
On Wed, Sep 28, 2016 at 10:24:07AM +0200, Juergen Gross wrote:
> On 27/09/16 17:57, Roger Pau Monne wrote:
> > Introduce a new %pB format specifier to print sizes (in bytes) in a
>
> Code suggests it is pZ.
Right, first implementation used pB, but then I realized this was already
used by Linux t
flight 101183 xen-unstable-smoke real [real]
http://logs.test-lab.xenproject.org/osstest/logs/101183/
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 1
For consumers not using (fully) C99-aware compilers, limit the number
of places where tweaking of the headers would be necessary: Introduce
and use xen_mk_ullong(), allowing its helper macro to be overridden at
once.
For now don't touch public/io/, which also has a few offenders.
The need to incl
On 28/09/16 12:56, Roger Pau Monne wrote:
> On Wed, Sep 28, 2016 at 10:24:07AM +0200, Juergen Gross wrote:
>> On 27/09/16 17:57, Roger Pau Monne wrote:
>>> Introduce a new %pB format specifier to print sizes (in bytes) in a
>> Code suggests it is pZ.
> Right, first implementation used pB, but then
>>> On 28.09.16 at 11:42, wrote:
> It is not used by anything.
But that shouldn't be the only aspect. Are they also not useful for
anything?
> --- a/xen/common/tmem_control.c
> +++ b/xen/common/tmem_control.c
> @@ -103,9 +103,9 @@ static int tmemc_list_client(struct client *c,
> tmem_cli_va_par
>>> On 28.09.16 at 11:42, wrote:
> This config is not defined anywhere but it makes it way
> easier to figure out what code to deal with.
And the functionality thus removed is
- dead code,
- broken,
- anything else?
> --- a/xen/common/tmem.c
> +++ b/xen/common/tmem.c
> @@ -71,10 +71,14 @@ struct
>>> On 28.09.16 at 11:42, wrote:
> . which is actually dependent on CONFIG_TMEM_DEDUP
Same comments as for patch 3.
Jan
___
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
On Tue, Sep 27, 2016 at 05:56:55PM +0200, Roger Pau Monne wrote:
> Hello,
>
> This is the first "complete" PVHv2 implementation in the sense that
> it has feature parity with classic PVH Dom0. It is still very experimental,
> but I've managed to boot it on all Intel boxes I've tried (OK, only 3 so
>>> On 28.09.16 at 11:42, wrote:
> Couple of reasons:
> - It can lead to security issues (see row-hammer, KSM and such
>attacks).
> - Code is quite complex.
> - Deduplication is good if the pages themselves are the same
>but that is hardly guaranteed.
> - We got some gains (if pages ar
>>> On 28.09.16 at 11:42, wrote:
> --- a/xen/include/public/sysctl.h
> +++ b/xen/include/public/sysctl.h
> @@ -785,6 +785,20 @@ struct tmem_handle {
> xen_tmem_oid_t oid;
> };
>
> +struct tmem_client {
I guess there's a reason for this to get added to the public header,
but if it really i
>>> On 28.09.16 at 11:42, wrote:
> Hypervisor wise we had to do a bit of casting. The
> 'struct xen_sysctl_tmem_op'->buf variable is a pointer to
> char. Casting that using the guest_handle_cast to a structure
> (struct tmem_client) does not work. Hence we first have to
> cast it a void and then t
On 09/28/2016 06:38 AM, Jan Beulich wrote:
On 27.09.16 at 16:46, wrote:
>> --- a/tools/firmware/hvmloader/acpi/Makefile
>> +++ b/tools/firmware/hvmloader/acpi/Makefile
>> @@ -17,7 +17,8 @@
>> XEN_ROOT = $(CURDIR)/../../../..
>> include $(XEN_ROOT)/tools/firmware/Rules.mk
>>
>> -C_SRC = bu
On Wed, Sep 28, 2016 at 08:53:41AM -0400, Boris Ostrovsky wrote:
> On 09/28/2016 06:38 AM, Jan Beulich wrote:
> On 27.09.16 at 16:46, wrote:
> >> --- a/tools/firmware/hvmloader/acpi/Makefile
> >> +++ b/tools/firmware/hvmloader/acpi/Makefile
> >> @@ -17,7 +17,8 @@
> >> XEN_ROOT = $(CURDIR)/..
>>> On 28.09.16 at 11:42, wrote:
> Note: We still have to do this awkward 'guest_handle_cast'
> otherwise it will not compile on ARM - which defines _two_
> of these macros (__guest_handle_64_xen_sysctl_tmem_client_t
> and __guest_handle_xen_sysctl_tmem_client_t). And if cast is
> not used then a
>>> On 28.09.16 at 11:42, wrote:
> --- a/xen/common/tmem_control.c
> +++ b/xen/common/tmem_control.c
> @@ -270,6 +270,12 @@ static int __tmemc_set_var(struct client *client,
> if ( copy_from_guest(&info, buf, 1) )
> return -EFAULT;
>
> +if ( info.version != TMEM_SPEC_VERSION )
>>> On 28.09.16 at 11:42, wrote:
> case XEN_SYSCTL_TMEM_OP_RESTORE_BEGIN:
> if ( client == NULL && (client = client_create(cli_id)) != NULL )
> -return 1;
> +rc = 0;
> +else
> +rc = -EEXIST;
> break;
I don't think the same error c
On 09/28/2016 08:53 AM, Wei Liu wrote:
> On Wed, Sep 28, 2016 at 08:53:41AM -0400, Boris Ostrovsky wrote:
>> On 09/28/2016 06:38 AM, Jan Beulich wrote:
>> On 27.09.16 at 16:46, wrote:
--- a/tools/firmware/hvmloader/acpi/Makefile
+++ b/tools/firmware/hvmloader/acpi/Makefile
@@ -1
On 2016/9/28 3:00, Wei Liu wrote:
On Tue, Sep 27, 2016 at 11:43:38AM -0700, Shannon Zhao wrote:
On 2016/9/27 9:35, Wei Liu wrote:
On Tue, Sep 27, 2016 at 09:01:00AM -0700, Shannon Zhao wrote:
On 2016/9/27 2:41, Wei Liu wrote:
On Mon, Sep 26, 2016 at 02:54:55PM -0700, Shannon Zhao wrote:
>>> On 28.09.16 at 11:42, wrote:
> @@ -1592,6 +1594,7 @@ static int do_tmem_new_pool(domid_t this_cli_id,
>
> out:
> tmem_client_info("pool_id=%d\n", d_poolid);
> +client->info.nr_pools ++;
Stray blank?
> --- a/xen/common/tmem_control.c
> +++ b/xen/common/tmem_control.c
> @@ -276,6 +
On Wed, Sep 28, 2016 at 06:11:53AM -0700, Shannon Zhao wrote:
>
>
> On 2016/9/28 3:00, Wei Liu wrote:
> >On Tue, Sep 27, 2016 at 11:43:38AM -0700, Shannon Zhao wrote:
> >>
> >>
> >>On 2016/9/27 9:35, Wei Liu wrote:
> >>>On Tue, Sep 27, 2016 at 09:01:00AM -0700, Shannon Zhao wrote:
>
>
>
On 28/09/16 13:00, Jan Beulich wrote:
> For consumers not using (fully) C99-aware compilers, limit the number
> of places where tweaking of the headers would be necessary: Introduce
> and use xen_mk_ullong(), allowing its helper macro to be overridden at
> once.
I'm just curious: Did you actually
flight 101185 xtf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/101185/
Perfect :-)
All tests in this flight passed as required
version targeted for testing:
xtf 962dd4b225c17ae48d36a978602b245dbf77bec5
baseline version:
xtf efe6103963c6756f5c7b77
This run is configured for baseline tests only.
flight 67778 ovmf real [real]
http://osstest.xs.citrite.net/~osstest/testlogs/logs/67778/
Perfect :-)
All tests in this flight passed as required
version targeted for testing:
ovmf ab970515d2c6ec657fceab0ce571054bb43a22f2
baseline v
>>> On 28.09.16 at 15:31, wrote:
> On 28/09/16 13:00, Jan Beulich wrote:
>> For consumers not using (fully) C99-aware compilers, limit the number
>> of places where tweaking of the headers would be necessary: Introduce
>> and use xen_mk_ullong(), allowing its helper macro to be overridden at
>> on
1 - 100 of 225 matches
Mail list logo
