On 10.08.2022 14:13, Andrew Cooper wrote:
> On 09/08/2022 16:50, Jan Beulich wrote:
>> When passed -1, the function (taking a u16) will look for segment
>> 0x, which might exist. If it exists, we may find (return) the wrong
>> device.
>>
>> Signed-off-by: Jan Beulich
>> ---
>> An alternative w
flight 172366 qemu-mainline real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172366/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-amd64-libvirt 6 libvirt-buildfail REGR. vs. 172123
build-i386-libvir
On 11.08.22 00:55, Marek Marczykowski-Górecki wrote:
On Thu, Jun 30, 2022 at 07:31:38AM +0200, Juergen Gross wrote:
On 30.06.22 05:32, SK, SivaSangeetha (Siva Sangeetha) wrote:
[AMD Official Use Only - General]
+team
-Original Message-
From: Stefano Stabellini
Sent: Thursday, June 30
There is a couple of tricks we can do with io_uring to improve ubuf_info
refcounting. First, we ammortise reference grabbing and then give them
away to the network layer, which is implemented in 8 and 11. Also, we
don't need need additional pinning for TCP, which is removed by 7.
1-4 are needed be
Some msg_ubuf providers like io_uring can keep elaborated ubuf_info
reference batching and caching, so it will be of benefit to let the
network layer to optionally steal some of the cached refs.
Add UARGFL_GIFT_REF, if set the caller has at least one extra reference
that it can gift away. If the n
struct ubuf_info will be changed, use ubuf_info_msgzc instead.
Signed-off-by: Pavel Begunkov
---
drivers/vhost/net.c | 17 +
1 file changed, 9 insertions(+), 8 deletions(-)
diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
index 68e4ecd1cc0e..9b616536dd9e 100644
--- a/drive
ubuf_info::flags contains SKBFL_* flags that we copy into skbs, change
the field name to stress that it keeps skb flags.
Signed-off-by: Pavel Begunkov
---
include/linux/skbuff.h | 4 ++--
io_uring/notif.c | 2 +-
net/core/skbuff.c | 2 +-
3 files changed, 4 insertions(+), 4 deletions(
struct ubuf_info will be changed, use ubuf_info_msgzc instead.
Signed-off-by: Pavel Begunkov
---
drivers/net/xen-netback/common.h| 2 +-
drivers/net/xen-netback/interface.c | 4 ++--
drivers/net/xen-netback/netback.c | 7 ---
3 files changed, 7 insertions(+), 6 deletions(-)
diff --git
We're going to split struct ubuf_info and leave there only
mandatory fields. Users are free to extend it. Add struct
ubuf_info_msgzc, which will be an extended version for MSG_ZEROCOPY and
some other users. It duplicates of struct ubuf_info for now and will be
removed in a couple of patches.
Signe
There are already skb_flags in ubuf_info, which enhancing skbs. Also add
flags controlling ubuf_info, mainly to hint about various referencing
aspects of it, which will be introduced in later patches.
Signed-off-by: Pavel Begunkov
---
include/linux/skbuff.h | 1 +
io_uring/notif.c | 1 +
n
We always keep references to active notifications and drop them only
when we flush, so they're always pinned during sock_sendmsg() and we can
add UARGFL_CALLER_PINNED.
Signed-off-by: Pavel Begunkov
---
io_uring/notif.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/io_uring/
Add a helper for dropping notification references during flush. It's a
preparation patch, currently it's only one master ref, but we're going
to add ref caching.
Signed-off-by: Pavel Begunkov
---
io_uring/notif.c | 14 +-
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/
Add UARGFL_CALLER_PINNED letting protocols know that the caller holds a
reference to the ubuf_info and so it doesn't need additional refcounting
for purposes of keeping it alive. With that TCP can save a refcount
put/get pair per send when used with ->msg_ubuf.
Signed-off-by: Pavel Begunkov
---
We can benefit from a smaller struct ubuf_info, so leave only mandatory
fields and let users to decide how they want to extend it. Convert
MSG_ZEROCOPY to struct ubuf_info_msgzc and remove duplicated fields.
This reduces the size from 48 bytes to just 16.
Signed-off-by: Pavel Begunkov
---
includ
Cache some active notifier references at the io_uring side and get them
in batches, so the ammortised cost is low. Then these references can be
given away to the network layer using UARGFL_GIFT_REF.
Signed-off-by: Pavel Begunkov
---
io_uring/net.c | 8 +++-
io_uring/notif.c | 6 --
i
flight 172375 libvirt real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172375/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-arm64-libvirt 6 libvirt-buildfail REGR. vs. 151777
build-armhf-libvirt
flight 172378 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172378/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-i386-libvirt6 libvirt-buildfail REGR. vs. 172136
build-amd64-libvirt
Hi Jan,
> On 11 Aug 2022, at 8:02 am, Jan Beulich wrote:
>
> On 10.08.2022 14:13, Andrew Cooper wrote:
>> On 09/08/2022 16:50, Jan Beulich wrote:
>>> When passed -1, the function (taking a u16) will look for segment
>>> 0x, which might exist. If it exists, we may find (return) the wrong
>>>
These functions may be called by init_amd() after the _init functions
have been purged during CPU hotplug or PV shim boot so drop the _init.
Signed-off-by: Ross Lagerwall
---
xen/arch/x86/cpu/common.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/xen/arch/x86/cpu/comm
On 11/08/2022 11:17, Ross Lagerwall wrote:
> These functions may be called by init_amd() after the _init functions
> have been purged during CPU hotplug or PV shim boot so drop the _init.
>
> Signed-off-by: Ross Lagerwall
Hmm. That's a bug in init_amd() I'd say. These really shouldn't be
used a
On 11/08/2022 11:30, Ross Lagerwall wrote:
>> From: Andrew Cooper
>> Sent: Thursday, August 11, 2022 11:21 AM
>> To: Ross Lagerwall ;
>> xen-devel@lists.xenproject.org
>> Cc: Jan Beulich ; Roger Pau Monne ;
>> Wei Liu
>> Subject: Re: [PATCH] x86/cpu: Drop _init from *_cpu_cap functions
>>
>
The last "wildcard" use of either function went away with f591755823a7
("IOMMU/PCI: don't let domain cleanup continue when device de-assignment
failed"). Don't allow them to be called this way anymore. Besides
simplifying the code this also fixes two bugs:
1) When seg != -1, the outer loops should
Rename the latter, subsuming the functionality of the former when passed
NULL as first argument.
Since this requires touching all call sites anyway, take the opportunity
and fold the remaining three parameters into a single pci_sbdf_t one.
No functional change intended. In particular the locking
Fold the three parameters into a single pci_sbdf_t one.
No functional change intended, despite the "(8 - stride)" ->
"stride" replacement (not really sure why it was written the more
complicated way originally).
Signed-off-by: Jan Beulich
---
v2: New.
--- a/xen/drivers/passthrough/amd/iommu_ini
The two previously submitted bug fixes can actually be had as a side
effect of eliminating a bogus feature, the last use of which had
disappeared a while ago. Further cleanup then follows along the
lines of what had also been discussed in the context of the earlier
attempts.
1: simplify (and thus
On 09/08/2022 18:00, Andrew Cooper wrote:
> Xen happens to be not vulnerable to PBRSB, but it turns out we can improve the
> performance on all eIBRS systems.
>
> Andrew Cooper (2):
> x86/spec-ctrl: Enumeration for PBRSB_NO
> x86/spec-ctrl: Reduce HVM RSB overhead where possible
Network perf t
On 11.08.2022 12:34, Andrew Cooper wrote:
> On 11/08/2022 11:30, Ross Lagerwall wrote:
>>> From: Andrew Cooper
>>> Sent: Thursday, August 11, 2022 11:21 AM
>>> To: Ross Lagerwall ;
>>> xen-devel@lists.xenproject.org
>>> Cc: Jan Beulich ; Roger Pau Monne
>>> ; Wei Liu
>>> Subject: Re: [PATCH] x
This should only be called for the boot CPU to avoid calling _init code
after it has been unloaded.
Fixes: 062868a5a8b4 ("x86/amd: Work around CLFLUSH ordering on older parts")
Signed-off-by: Ross Lagerwall
---
This replaces "x86/cpu: Drop _init from *_cpu_cap functions".
xen/arch/x86/cpu/amd.
flight 172369 linux-linus real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172369/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-amd64-libvirt 6 libvirt-buildfail REGR. vs. 172133
build-i386-libvirt
flight 172381 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172381/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-i386-libvirt6 libvirt-buildfail REGR. vs. 172136
build-amd64-libvirt
On Thu, Aug 11, 2022 at 2:51 AM Jan Beulich wrote:
>
> On 10.08.2022 21:29, Jason Andryuk wrote:
> > --- a/CHANGELOG.md
> > +++ b/CHANGELOG.md
> > @@ -6,6 +6,9 @@ The format is based on [Keep a
> > Changelog](https://keepachangelog.com/en/1.0.0/)
> >
> > ## [unstable
> > UNRELEASED](https://xen
The double `the' is duplicated in the comment, remove one.
Signed-off-by: Jason Wang
---
drivers/xen/xen-pciback/pciback_ops.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/xen/xen-pciback/pciback_ops.c
b/drivers/xen/xen-pciback/pciback_ops.c
index 3fbc21466a93..84
Commit 68f5aac012b9 ("build: suppress future GNU ld warning about RWX
load segments") didn't quite cover all the cases: Apparently I missed
ones in the building of 32-bit helper objects because of only looking at
incremental builds (where those wouldn't normally be re-built). Clone
the workaround t
It's not clear why for x86-64 a different approach was used than the
(shorter) one x86-32 has been using. Move the LDFLAGS_DIRECT_*BSD
setting to the respective OS files and reuse x86-32's approach for
x86-64.
Signed-off-by: Jan Beulich
---
I was actually inclined to go further and rename the var
Hi Jason,
> -Original Message-
> From: Jason Andryuk
> For v3, I'll move to the existing section while keeping Henry's Ack.
Thank you. Keeping the Ack is ok to me.
Kind regards,
Henry
>
> Regards,
> Jason
On 11.08.2022 12:59, Ross Lagerwall wrote:
> This should only be called for the boot CPU to avoid calling _init code
> after it has been unloaded.
>
> Fixes: 062868a5a8b4 ("x86/amd: Work around CLFLUSH ordering on older parts")
> Signed-off-by: Ross Lagerwall
Reviewed-by: Jan Beulich
albeit ...
flight 172374 xen-unstable real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172374/
Failures :-/ but no regressions.
Tests which did not succeed, but are not blocking:
test-amd64-i386-libvirt-qemuu-debianhvm-amd64-xsm 1 build-check(1) blocked n/a
test-amd64-i386-libvirt-raw 1 buil
On Tue, Aug 09, 2022 at 08:17:49AM -0400, Jason Andryuk wrote:
> On Mon, Aug 8, 2022 at 7:06 AM Anthony PERARD
> wrote:
> > diff --git a/docs/man/xl.cfg.5.pod.in b/docs/man/xl.cfg.5.pod.in
> > index 6d98d73d76..b2901e04cf 100644
> > --- a/docs/man/xl.cfg.5.pod.in
> > +++ b/docs/man/xl.cfg.5.pod.i
flight 172382 xen-unstable-smoke real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172382/
Failures :-/ but no regressions.
Tests which did not succeed, but are not blocking:
test-amd64-amd64-libvirt 1 build-check(1) blocked n/a
build-amd64-libvirt 6 lib
On 11/08/2022 11:51, Jan Beulich wrote:
> The last "wildcard" use of either function went away with f591755823a7
> ("IOMMU/PCI: don't let domain cleanup continue when device de-assignment
> failed"). Don't allow them to be called this way anymore. Besides
> simplifying the code this also fixes two
On 11.08.2022 15:11, Andrew Cooper wrote:
> On 11/08/2022 11:51, Jan Beulich wrote:
>> The last "wildcard" use of either function went away with f591755823a7
>> ("IOMMU/PCI: don't let domain cleanup continue when device de-assignment
>> failed"). Don't allow them to be called this way anymore. Besi
On 11/08/2022 11:52, Jan Beulich wrote:
> Rename the latter, subsuming the functionality of the former when passed
> NULL as first argument.
>
> Since this requires touching all call sites anyway, take the opportunity
> and fold the remaining three parameters into a single pci_sbdf_t one.
>
> No fu
On 11.08.2022 15:21, Andrew Cooper wrote:
> On 11/08/2022 11:52, Jan Beulich wrote:
>> --- a/xen/arch/x86/irq.c
>> +++ b/xen/arch/x86/irq.c
>> @@ -2162,7 +2162,7 @@ int map_domain_pirq(
>> if ( !cpu_has_apic )
>> goto done;
>>
>> -pdev = pci_get_pdev_by_domain(d, msi
On 11/08/2022 11:52, Jan Beulich wrote:
> Fold the three parameters into a single pci_sbdf_t one.
>
> No functional change intended, despite the "(8 - stride)" ->
> "stride" replacement (not really sure why it was written the more
> complicated way originally).
>
> Signed-off-by: Jan Beulich
Revi
Hi,
On Thu, Jul 28, 2022 at 9:41 PM Julien Grall wrote:
>
> Hi,
>
> On 27/07/2022 07:33, Jens Wiklander wrote:
> > On Fri, Jul 8, 2022 at 9:54 PM Julien Grall wrote:
> >>> +unsigned int n;
> >>> +unsigned int m;
> >>> +p2m_type_t t;
> >>> +uint64_t addr;
> >>> +
> >>> +for (
Hi Bertrand, Rahul,
On Fri, Aug 05, 2022 at 12:05:23PM +, Bertrand Marquis wrote:
> > On 5 Aug 2022, at 12:44, Rahul Singh wrote:
[...]
> >> Looked into the code, the GICv3 driver tries to create persistent
> >> reservations for pending pages, and the persistent reservation table
> >> can b
flight 172379 qemu-mainline real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172379/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-amd64-libvirt 6 libvirt-buildfail REGR. vs. 172123
build-i386-libvir
Hi Leon,
> On 11 Aug 2022, at 15:17, Leo Yan wrote:
>
> Hi Bertrand, Rahul,
>
> On Fri, Aug 05, 2022 at 12:05:23PM +, Bertrand Marquis wrote:
>>> On 5 Aug 2022, at 12:44, Rahul Singh wrote:
>
> [...]
>
Looked into the code, the GICv3 driver tries to create persistent
reservatio
On Thu, 11 Aug 2022 at 16:59, Bertrand Marquis wrote:
>
> Hi Leon,
>
> > On 11 Aug 2022, at 15:17, Leo Yan wrote:
> >
> > Hi Bertrand, Rahul,
> >
> > On Fri, Aug 05, 2022 at 12:05:23PM +, Bertrand Marquis wrote:
> >>> On 5 Aug 2022, at 12:44, Rahul Singh wrote:
> >
> > [...]
> >
> Looke
On 11/08/2022 14:26, Jan Beulich wrote:
> On 11.08.2022 15:21, Andrew Cooper wrote:
>> On 11/08/2022 11:52, Jan Beulich wrote:
>>> --- a/xen/arch/x86/irq.c
>>> +++ b/xen/arch/x86/irq.c
>>> @@ -2162,7 +2162,7 @@ int map_domain_pirq(
>>> if ( !cpu_has_apic )
>>> goto done;
>>>
On 11/08/2022 13:12, Jan Beulich wrote:
> Commit 68f5aac012b9 ("build: suppress future GNU ld warning about RWX
> load segments") didn't quite cover all the cases: Apparently I missed
> ones in the building of 32-bit helper objects because of only looking at
> incremental builds (where those wouldn
On 11/08/2022 13:14, Jan Beulich wrote:
> It's not clear why for x86-64 a different approach was used than the
> (shorter) one x86-32 has been using. Move the LDFLAGS_DIRECT_*BSD
> setting to the respective OS files and reuse x86-32's approach for
> x86-64.
>
> Signed-off-by: Jan Beulich
Acked-by
Hi,
> On 11 Aug 2022, at 16:05, Ard Biesheuvel wrote:
>
> On Thu, 11 Aug 2022 at 16:59, Bertrand Marquis
> wrote:
>>
>> Hi Leon,
>>
>>> On 11 Aug 2022, at 15:17, Leo Yan wrote:
>>>
>>> Hi Bertrand, Rahul,
>>>
>>> On Fri, Aug 05, 2022 at 12:05:23PM +, Bertrand Marquis wrote:
> On 5
On 11.08.2022 17:09, Andrew Cooper wrote:
> On 11/08/2022 14:26, Jan Beulich wrote:
>> On 11.08.2022 15:21, Andrew Cooper wrote:
>>> On 11/08/2022 11:52, Jan Beulich wrote:
--- a/xen/arch/x86/irq.c
+++ b/xen/arch/x86/irq.c
@@ -2162,7 +2162,7 @@ int map_domain_pirq(
if (
When devices are deassigned/assigned, SMMU global fault is observed
because SMEs are freed in detach function and not allocated again when
the device is assigned back to the guest.
Don't free the SMEs when devices are deassigned, set the s2cr to type
fault. This way the SMMU will generate a fault
flight 172386 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172386/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-i386-libvirt6 libvirt-buildfail REGR. vs. 172136
build-amd64-libvirt
Hi Jan,
> On 11 Aug 2022, at 11:51 am, Jan Beulich wrote:
>
> The last "wildcard" use of either function went away with f591755823a7
> ("IOMMU/PCI: don't let domain cleanup continue when device de-assignment
> failed"). Don't allow them to be called this way anymore. Besides
> simplifying the co
Hi Jan,
> On 11 Aug 2022, at 11:52 am, Jan Beulich wrote:
>
> Fold the three parameters into a single pci_sbdf_t one.
>
> No functional change intended, despite the "(8 - stride)" ->
> "stride" replacement (not really sure why it was written the more
> complicated way originally).
>
> Signed-o
Hi Jan,
> On 11 Aug 2022, at 11:52 am, Jan Beulich wrote:
>
> Rename the latter, subsuming the functionality of the former when passed
> NULL as first argument.
>
> Since this requires touching all call sites anyway, take the opportunity
> and fold the remaining three parameters into a single p
This reorders the fields in msi_info, but removes all the under-the-hood
parameter shuffling required to call pci_get_pdev().
No functional change.
Signed-off-by: Andrew Cooper
---
CC: Jan Beulich
CC: Roger Pau Monné
CC: Wei Liu
---
xen/arch/x86/hvm/vmsi.c| 4 +---
xen/arch/x86/incl
Add "xentop" to "TARGETS" because this variable will be useful later.
Always define all the targets, even when configured with
--disable-monitor, instead don't visit the subdirectory.
This mean xentop/ isn't visited anymore during "make clean" that's how
most other subdirs in the tools/ works.
Al
Patch series available in this git branch:
https://xenbits.xen.org/git-http/people/aperard/xen-unstable.git
br.toolstack-build-system-v4
Changes in v4:
- several new patches
- some changes to other patches listed in their changelogs
Changes in v3:
- rebased
- several new patches, starting with 1
gdbsx/:
- Make use of subdir facility for the "clean" target.
- No need to remove the *.a, they aren't in this dir.
- Avoid calling "distclean" in subdirs as "distclean" targets do only
call "clean", and the "clean" also runs "clean" in subdirs.
- Avoid the need to make "gx_all.a" and "
Signed-off-by: Anthony PERARD
Reviewed-by: Juergen Gross
Reviewed-by: Bertrand Marquis
---
tools/libs/libs.mk | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/tools/libs/libs.mk b/tools/libs/libs.mk
index 58d8166b09..e02f91f95e 100644
--- a/tools/libs/libs.mk
+++ b/to
Remove the need for "fs-*" targets by creating a "common.mk" which
have flags that are common to libfsimage/common/ and the other
libfsimages/*/ directories.
In common.mk, make $(PIC_OBJS) a recursively expanded variable so it
doesn't matter where $(LIB_SRCS-y) is defined, and remove the extra
$(P
Remove "build" targets.
Use simply expanded variables when recursively expanded variable
aren't needed. (Use ":=" instead of "=".)
Don't check if a directory already exist when installing, just create
it.
Fix $(HOTPLUGPATH), it shouldn't have any double-quote.
Some reindentation.
FreeBSD, "hot
Rework dependencies of all objects. We don't need to add dependencies
for headers that $(CC) is capable of generating, we only need to
include $(DEPS_INCLUDE). Some dependencies are still needed so make
knows to generate symlinks for them.
We remove the use of "vpath" for cpuid.c. While it works f
Signed-off-by: Anthony PERARD
Reviewed-by: Henry Wang
---
tools/flask/utils/Makefile | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tools/flask/utils/Makefile b/tools/flask/utils/Makefile
index db567b13dc..6be134142a 100644
--- a/tools/flask/utils/Makefile
+++ b/t
- Rename $(SRCS) to $(OBJS-y), we don't need to collect sources.
- Rename $(IBINS) to $(TARGETS)
- Stop cleaning "xen" and non-set variable $(LIB).
Signed-off-by: Anthony PERARD
Reviewed-by: Luca Fancellu
---
tools/xenpaging/Makefile | 24
1 file changed, 12 insertions(
Signed-off-by: Anthony PERARD
Reviewed-by: Juergen Gross
Reviewed-by: Bertrand Marquis
---
Notes:
v2:
- move new .gitignore entries to the one in tools/libs/
.gitignore| 26 --
tools/libs/.gitignore | 2 ++
2 files changed, 2 insertions(+), 26 dele
Setup proper dependencies with libacpi so we don't need to run "make
hvmloader" in the "all" target. ("build.o" new prerequisite isn't
exactly proper but a side effect of building the $(DSDT_FILES) is to
generate the "ssdt_*.h" needed by "build.o".)
Make use if "-iquote" instead of a plain "-I".
There is no need to call the "headers.chk" target when it isn't
wanted, so it never need to be .PHONY.
Also, there is no more reason to separate the prerequisites from the
recipe.
Signed-off-by: Anthony PERARD
Reviewed-by: Juergen Gross
Reviewed-by: Bertrand Marquis
---
tools/libs/libs.mk | 1
We can simply use $(PKG_CONFIG) to set the parameters, and add it to
$(TARGETS) as necessary.
Signed-off-by: Anthony PERARD
Reviewed-by: Juergen Gross
Reviewed-by: Bertrand Marquis
---
tools/libs/libs.mk | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/tools/libs
This patch introduce a new macro $(xenlibs-dependencies,) to generate
a list of all the xen library that a library is list against, and they
are listed only once. We use the side effect of $(sort ) which remove
duplicates.
This is used by another macro $(xenlibs-rpath,) which is to replace
$(SHDEP
No need for $(AUTOSRCS), GNU make can generate them as needed when
trying to build them as needed when trying to build the object. Also,
those two AUTOSRCS don't need to be a prerequisite of "all". As for
the "clean" target, those two files are already removed via "_*.c".
We don't need $(AUTOINCS)
This doesn't matter yet but it will when for example the script will
be run from tools/ to generate files tools/libs/light/.
Signed-off-by: Anthony PERARD
---
Notes:
v4:
- new patch
tools/libs/light/gentypes.py | 9 ++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git
Remove -I. from CFLAGS, it isn't necessary.
Removed $(AUTOSRCS), it isn't used.
Signed-off-by: Anthony PERARD
Reviewed-by: Juergen Gross
Reviewed-by: Bertrand Marquis
---
tools/libs/util/Makefile | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/tools/libs/util/Makefile b/
Rework "xen-xsm" rules to not have to change directory to run
mkflask.sh, and store mkflask.sh path in a var, and use a full path
for FLASK_H_DEPEND, and output directory is made relative.
Rename "all-y" target to a more descriptive "xen/lib/x86/all".
Removed the "dist" target which was the only
$(SHLIB_LDFLAGS) is used by more targets that the single targets that
except it (libxenfoo.so.X.Y). There is also some dynamic libraries in
stats/ that uses $(SHLIB_LDFLAGS) (even if those are never built), and
there's libxenlight_test.so which doesn't needs a version script.
Also, libxenlight_tes
Patch "tools: Add -Werror by default to all tools/" have added
"-Werror" to CFLAGS in tools/Rules.mk, remove it from every other
makefiles as it is now duplicated.
Signed-off-by: Anthony PERARD
Acked-by: Christian Lindig
---
tools/console/client/Makefile | 1 -
tools/console/daemon/Makefile
Currently, a rebuild of libxl will always rebuild for example
"build.o". This is because the target depends on "acpi" which never
exist. So instead we will have "build.o" have as prerequisites targets
that are actually generated by "acpi", that is $(DSDT_FILES-y).
While "dsdt_*.c" isn't really a d
Remove "build" targets.
Use "$(TARGETS)" to list binary to be built.
Cleanup "clean" rule.
Also drop conditional install of $(BIN) and $(LIBBIN) as those two
variables are now always populated.
Signed-off-by: Anthony PERARD
---
Notes:
v4:
- also use := for BIN, SBIN, LIBBIN, SCRIPTS
Rework "libacpi.h" include in "libxl_x86_acpi.c" as to be more
selective about the include path and only add "tools/libacpi/". Also
"libxl_dom.c" don't use "libacpi.h" anymore. Use "-iquote" for libacpi
headers.
Get rid of the weird "$(eval stem =" in the middle of a recipe and use
a make automati
gengotypes.py creates both "types.gen.go" and "helpers.gen.go", but
make can start gengotypes.py twice. Rework the rules so that
gengotypes.py is executed only once.
Also, add the ability to provide a path to tell gengotypes.py where to
put the files. This doesn't matter yet but it will when for e
This avoid the need to open-coding the list of flags needed to link
with an in-tree Xen library when using -lxen*.
Signed-off-by: Anthony PERARD
Reviewed-by: Henry Wang
---
tools/golang/xenlight/Makefile | 2 +-
tools/Rules.mk | 8
2 files changed, 9 insertions(+), 1 de
"$DIR" might not be a full path and it might not have `pwd` as ".."
directory. So use `cd -` to undo the first `cd` command.
Also, use `basename` to make a symbolic link with a relative path.
This doesn't matter yet but it will when for example the commands to
clone OVMF is been run from tools/ r
Use $(TARGETS) to collect targets. Use := for the first target instead
of +=.
Collect library to link against in $(LDLIBS).
Remove extra "-f" flags that is already part of $(RM).
Signed-off-by: Anthony PERARD
Reviewed-by: Luca Fancellu
Tested-by: Luca Fancellu
---
Notes:
v4:
- use := f
When there isn't a version-script for a shared library (like for
unstable libs), we create one based on the current Xen version. But
that version-script became out-of-date as soon as Xen's version
changes and make as no way to regenerate the version-script on
rebuild.
In order to allow regeneratin
Instead of moving the public "_libxl_*.h" headers, we make a copy to
the destination so that make doesn't try to remake the targets
"_libxl_*.h" in libs/light/ again.
A new .PRECIOUS target is added to tell make to not deletes the
intermediate targets generated by "gentypes.py".
Signed-off-by: An
This can be used when linking against multiple in-tree Xen libraries,
and avoid duplicated flags. It can be used instead of multiple
$(LDLIBS_libxen*).
For now, replace the open-coding in libs.mk.
The macro $(xenlibs-libs, ) will be useful later when only the path to
the libraries is wanted (e.g.
Signed-off-by: Anthony PERARD
---
Notes:
v4:
- new patch
tools/configure.ac | 1 +
tools/hotplug/common/Makefile | 10 ++
tools/configure| 3 ++-
tools/hotplug/common/hotplugpath.sh.in | 16
4 files changed,
Signed-off-by: Anthony PERARD
Reviewed-by: Luca Fancellu
---
tools/console/client/Makefile | 3 +--
tools/console/daemon/Makefile | 6 +-
2 files changed, 2 insertions(+), 7 deletions(-)
diff --git a/tools/console/client/Makefile b/tools/console/client/Makefile
index 44176c6d93..e2f2554f92
And provide an option to ./configure to disable it.
A follow-up patch will remove -Werror from every other Makefile in
tools/. ("tools: Remove -Werror everywhere else")
Signed-off-by: Anthony PERARD
Reviewed-by: Luca Fancellu
---
tools/configure.ac | 1 +
tools/Rules.mk | 4
config/
On 10/08/2022 09:00, Jan Beulich wrote:
> On 09.08.2022 19:00, Andrew Cooper wrote:
>> --- a/xen/arch/x86/hvm/vmx/entry.S
>> +++ b/xen/arch/x86/hvm/vmx/entry.S
>> @@ -44,6 +44,7 @@ ENTRY(vmx_asm_vmexit_handler)
>> .endm
>> ALTERNATIVE "", restore_spec_ctrl, X86_FEATURE_SC_MSR_HVM
The linux_xen template can put multiple initrd files in the
same multiboot[2] module[2] command, which is against specs.
This causes ONLY the _first_ initrd file to be loaded; other
files just have filenames in a "cmdline" string of the first
initrd file and are NOT loaded.
Fix this by inserting
Adding the xen-devel list as requested in grub-devel [0]:
On Thu, Aug 11, 2022 at 1:51 PM Daniel Kiper wrote:
> In general patches LGTM. However, I would prefer to hear an opinion from Xen
> folks too. So, please repost the patch set and add
> xen-devel@lists.xenproject.org
> to the list of reci
The linux_xen template orders the "early" initrd file(s) _first_
(i.e., before the "real" initrd files) and that seems reasonable,
as microcode updates usually come first.
However, this usually breaks Linux boot with initrd under Xen
because Xen assumes the real initrd is the first multiboot[2]
mo
flight 172391 ovmf real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172391/
Regressions :-(
Tests which did not succeed and are blocking,
including tests which could not be run:
build-i386-libvirt6 libvirt-buildfail REGR. vs. 172136
build-amd64-libvirt
flight 172390 xen-unstable-smoke real [real]
http://logs.test-lab.xenproject.org/osstest/logs/172390/
Failures :-/ but no regressions.
Tests which did not succeed, but are not blocking:
test-amd64-amd64-libvirt 1 build-check(1) blocked n/a
build-amd64-libvirt 6 lib
One source of lost performance was that fact that to protect Xen from a
malicious guests, we had to flush the RAS.
It turns out that CET Shadow Stacks give us enough architectural guarantees to
construct a lower overhead mitigation, which keeps the RAS balanced for the
guest so their return perfor
1 - 100 of 113 matches
Mail list logo
