On 12.05.2016 05:48, Bharata B Rao wrote:
> Hi,
>
> This is v3 of "Core based CPU hotplug for PowerPC sPAPR". The hotplug
> semantics looks like this:
>
> (qemu) device_add POWER8E-spapr-cpu-core,id=core2,core=16[,threads=4]
> (qemu) device_add POWER8E_v2.1-spapr-cpu-core,id=core2,core=16[,thread
* Drop the old SysBus init function
* Call qemu_chr_add_handlers in the realize callback
* Use qdev chardev prop instead of qemu_char_get_next_serial
Signed-off-by: xiaoqiang zhao
---
hw/char/lm32_juart.c | 17 -
hw/lm32/lm32.h| 3 ++-
hw/lm32/lm32_boards.c | 5 +++--
On Tue, May 24, 2016 at 10:55:07AM -0700, Jianjun Duan wrote:
> A recursive structure has elements of the same type in itself. Currently
> we cannot directly transfer a QTAILQ instance, or any recursive
> structure such as lists in migration because of the limitation in the
> migration code. Here w
* Drop the old SysBus init function and use instance_init
* Call qemu_chr_add_handlers in the realize callback
* Use qdev chardev prop instead of qemu_char_get_next_serial
* Add etraxfs_ser_create function to create etraxfs serial device
Signed-off-by: xiaoqiang zhao
---
hw/char/etraxfs_ser.c
This patch set trys to QOM'ify hw/char files, see commit messages
for more details
Thanks Paolo for your suggestions.
Note:
* CRIS axis_dev88 broad related test is passed and looks ok.
* lm32 test cases by Michael is passed and looks good.
* lm32 milkymist need test by Michael.
Changes in v6
On Fri, May 13, 2016 at 04:24:53PM -0600, Alex Williamson wrote:
> On Fri, 13 May 2016 17:16:48 +1000
> Alexey Kardashevskiy wrote:
>
> > On 05/06/2016 08:39 AM, Alex Williamson wrote:
> > > On Wed, 4 May 2016 16:52:13 +1000
> > > Alexey Kardashevskiy wrote:
> > >
> > >> This postpones VFIO c
* Drop the old SysBus init function and use instance_init
* Call qemu_chr_add_handlers in the realize callback
Signed-off-by: xiaoqiang zhao
---
hw/char/escc.c | 30 +++---
1 file changed, 19 insertions(+), 11 deletions(-)
diff --git a/hw/char/escc.c b/hw/char/escc.c
ind
* Drop the old SysBus init function and use instance_init
* Call qemu_chr_add_handlers in the realize callback
* Use qdev chardev prop instead of qemu_char_get_next_serial
* Add lm32_uart_create function to create lm32 uart device
Tested-by: Michael Walle
Acked-by: Michael Walle
Signed-off-by: x
drop the qemu_char_get_next_serial and use chardev prop instead
Signed-off-by: xiaoqiang zhao
---
hw/char/milkymist-uart.c | 10 ++
hw/lm32/milkymist-hw.h | 4 +++-
hw/lm32/milkymist.c | 2 +-
3 files changed, 10 insertions(+), 6 deletions(-)
diff --git a/hw/char/milkymist-uart
On 05/24/2016 09:47 PM, Kevin Wolf wrote:
There is a single remaining user in qemu-img, which can be trivially
converted to using BlockJob.blk instead.
Signed-off-by: Kevin Wolf
Reviewed-by: Max Reitz
---
blockjob.c | 1 -
include/block/blockjob.h | 1 -
qemu-img.c
Hi, Alex
3. Stall any access to the device until resume is signaled.
The code below doesn't actually do this, mmaps are disabled, but that
just means any device access will use the slow path through QEMU. That
path is still fully enabled and so is config space access.
I will modify the cod
From: Riku Voipio
QEMU lists deprecated system call numbers in for Aarch64. These
are never enabled for Linux kernel, so don't define them in Qemu
either. Remove the ifdef around host_to_target_stat64 since
all architectures need it now.
Reviewed-by: Peter Maydell
Signed-off-by: Riku Voipio
--
** Project changed: qemu => magnum
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1585433
Title:
if docker-volume-size of baymodel lessthan 3, bay cann't create
Status in Magnum:
New
Bug descri
>>> Kevin Wolf 2016/5/23 星期一 下午 5:00 >>>
>Am 22.05.2016 um 11:55 hat Lin Ma geschrieben:
>> Currently, the output of 'info snapshots' show fully available snapshots.
>>
>> In my opinion there are 2 disadvantages:
>> 1. It's opaque, hides some snapshot information to users. It's not convenient
From: Timothy E Baldwin
Update the x86 main loop and sigreturn code:
* on TARGET_ERESTARTSYS, wind guest PC backwards to repeat syscall insn
* set all guest CPU state within signal.c code rather than passing it
back out as the "return code" from do_sigreturn()
* handle TARGET_QEMU_ESIGRETUR
From: Peter Maydell
Our implementation of shmat() and shmdt() for linux-user was
using "zero guest address" as its marker for "entry in the
shm_regions[] array is not in use". This meant that if the
guest did a shmdt(0) we would match on an unused array entry
and call page_set_flags() with both s
I assume you want this to be in the Magnum project, not qemu...?
** Project changed: qemu => magnum
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1585432
Title:
magnum coe-service-list returns er
Hi,
CPU: Cortex R5F
I have this instruction that invalidates the entire data cache
MCR p15, 0, r0, c15, c5, 0
http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0460c/Chdhgibd.html
This instruction generates undefined exception, and further debugging
showed it is because the co-pr
From: Timothy E Baldwin
Define TARGET_ERESTARTSYS; like the kernel, we will use this to
indicate that a guest system call should be restarted. We use
the same value the kernel does for this, 512.
Signed-off-by: Timothy Edward Baldwin
[PMM: split out from the patch which moves and renumbers
TAR
Hi,
Sorry ignore these "pull v2" mails, apparently I had some old files
lying around..
On Wed, May 25, 2016 at 08:27:38AM +0300, riku.voi...@linaro.org wrote:
> From: Laurent Vivier
>
> Set the default to the latest CPU version to have the
> largest set of available features.
>
> It is also re
From: Laurent Vivier
target_fd_trans is an array of "TargetFdTrans *": compute size
accordingly. Use g_renew() as proposed by Paolo.
Reported-by: Paolo Bonzini
Signed-off-by: Laurent Vivier
Signed-off-by: Riku Voipio
---
linux-user/syscall.c | 4 ++--
1 file changed, 2 insertions(+), 2 delet
On Tue, May 24, 2016 at 10:55:04AM -0700, Jianjun Duan wrote:
> There are possible racing situations involving hotplug events and
> guest migration. For cases where a hotplug event is migrated, or
> the guest is in the process of fetching device tree at the time of
> migration, we need to ensure th
From: Timothy E Baldwin
Some of the signal handling was a mess with a mixture of tabs and 8 space
indents.
Signed-off-by: Timothy Edward Baldwin
Message-id: 1441497448-32489-3-git-send-email-t.e.baldwi...@members.leeds.ac.uk
Reviewed-by: Peter Maydell
[PMM: just rebased]
Signed-off-by: Peter M
From: Peter Maydell
The function do_openat() is not consistent about whether it is
returning a host errno or a guest errno in case of failure.
Standardise on returning -1 with errno set (ie caller has
to call get_errno()).
Signed-off-by: Peter Maydell
Reported-by: Timothy Edward Baldwin
Signed
Hi,
Added enabling VFP co-processor in the bootrom emulation code. Works like a
charm.
Many thanks.
Best regards,
Karthik
On Tue, May 24, 2016 at 7:52 PM, Karthik wrote:
> Yeah, the micro had a secure boot ROM which we don't have access to.
> Probably it is enabled there. I'll check in the t
From: Timothy E Baldwin
Currently we define a QEMU-internal errno TARGET_QEMU_ESIGRETURN
only on the MIPS and PPC targets; move this to errno_defs.h
so it is available for all architectures, and renumber it to 513.
We pick 513 because this is safe from future use as a system call return
value: Li
From: Timothy E Baldwin
Check array bounds in host_to_target_errno() and target_to_host_errno().
Signed-off-by: Timothy Edward Baldwin
Message-id: 1441497448-32489-2-git-send-email-t.e.baldwi...@members.leeds.ac.uk
[PMM: Add a lower-bound check, use braces on if(), tweak commit message]
Signed-
From: Timothy E Baldwin
Update the 32-bit and 64-bit ARM main loop and sigreturn code:
* on TARGET_ERESTARTSYS, wind guest PC backwards to repeat syscall insn
* set all guest CPU state within signal.c code on sigreturn
* handle TARGET_QEMU_ESIGRETURN in the main loop as the indication
that
From: Lluís Vilanova
This fixes double-definitions in linux-user builds when using the UST
tracing backend (which indirectly includes the system's "syscall.h").
Signed-off-by: Lluís Vilanova
Reviewed-by: Peter Maydell
Signed-off-by: Riku Voipio
---
linux-user/aarch64/syscall.h | 1
From: Riku Voipio
Sync syscall numbers to match the linux v4.5-rc1 kernel.
Reviewed-by: Laurent Vivier
Signed-off-by: Riku Voipio
---
linux-user/aarch64/syscall_nr.h| 2 +-
linux-user/alpha/syscall_nr.h | 6 ++
linux-user/cris/syscall_nr.h | 24
l
From: Laurent Vivier
Set the default to the latest CPU version to have the
largest set of available features.
It is also really needed in little-endian mode because
POWER7 is not really supported in this mode and some distros
(at least debian) generate POWER8 code for their ppc64le target.
Fixe
From: Riku Voipio
The following changes since commit b0f6ef8915247f3230ffd9b71af9c3dadb6082c7:
Merge remote-tracking branch 'remotes/amit-virtio-rng/tags/rng-2.7-1' into
staging (2016-05-24 11:38:22 +0100)
are available in the git repository at:
git://git.linaro.org/people/riku.voipio/qem
Similar to commit df7b97ff, we are mishandling clients that
give an unaligned NBD_CMD_TRIM request, and potentially
trimming bytes that occur before their request; which in turn
can cause potential unintended data loss (unlikely in
practice, since most clients are sane and issue aligned trim
reques
On 05/25/2016 12:01 PM, Eric Blake wrote:
On 05/24/2016 09:51 PM, Changlong Xie wrote:
On 05/24/2016 09:47 PM, Kevin Wolf wrote:
+s->target = blk_new();
blk_new(errp);
Depends on Kevin's block/next branch, which currently includes:
commit 5d7dd50566a4f9786b95f49448f48fead0bb34d8
Author:
On 05/24/2016 09:51 PM, Changlong Xie wrote:
> On 05/24/2016 09:47 PM, Kevin Wolf wrote:
>> +s->base = blk_new();
> blk_new(errp);
>> +blk_insert_bs(s->base, base);
>> +
>> +s->top = blk_new();
> blk_new(errp);
Wrong. Even if it weren't for basing it on top of Kevin's branch which
remo
On 05/24/2016 09:51 PM, Changlong Xie wrote:
> On 05/24/2016 09:47 PM, Kevin Wolf wrote:
>> +s->target = blk_new();
> blk_new(errp);
Depends on Kevin's block/next branch, which currently includes:
commit 5d7dd50566a4f9786b95f49448f48fead0bb34d8
Author: Max Reitz
Date: Tue May 17 16:41:34 2
On 05/24/2016 09:47 PM, Kevin Wolf wrote:
+job->target = blk_new();
blk_new(errp);
On 05/24/2016 09:47 PM, Kevin Wolf wrote:
+s->target = blk_new();
blk_new(errp);
On 05/24/2016 09:47 PM, Kevin Wolf wrote:
+s->base = blk_new();
blk_new(errp);
+blk_insert_bs(s->base, base);
+
+s->top = blk_new();
blk_new(errp);
On 05/24/2016 09:47 PM, Kevin Wolf wrote:
+blk = blk_new();
blk_new(errp);
On Tue, 24 May 2016 13:49:12 +0300
"Michael S. Tsirkin" wrote:
> On Tue, Apr 26, 2016 at 08:48:15AM -0600, Alex Williamson wrote:
> > I think that means that if we want to switch from a
> > simple halt-on-error to a mechanism for the guest to handle recovery,
> > we need to disable access to the
Public bug reported:
magnum is running on centos7,
magnum baymodel-create --name k8sbaymodel5 --image-id fedora-23-atomic-20160405
--keypair-id testkey --external-network-id public --dns-nameserver 8.8.8.8
--flavor-id m1.small --coe kubernetes --docker-volume-size 5
magnum bay-create --name k
Public bug reported:
magnum running on centos7,
I didn't create any services on k8s cluster,
but I got the result as follows:
[root@localhost ~(keystone_admin)]# magnum coe-service-list --bay k8sbay3
ERROR: Field `ports[0][node_port]' cannot be None (HTTP 500) (Request-ID:
req-c66b68dd-5437-47
On 05/24/2016 07:47 AM, Kevin Wolf wrote:
> This changes the commit block job to use the job's BlockBackend for
> performing its I/O. job->bs isn't used by the commit code any more
> afterwards.
>
> Signed-off-by: Kevin Wolf
> ---
> block/commit.c | 53 +--
On 05/24/2016 07:47 AM, Kevin Wolf wrote:
> This changes the backup block job to use the job's BlockBackend for
> performing its I/O. job->bs isn't used by the backup code any more
> afterwards.
>
> Signed-off-by: Kevin Wolf
> ---
> block/backup.c| 46 +---
On 05/24/2016 07:47 AM, Kevin Wolf wrote:
> This changes the mirror block job to use the job's BlockBackend for
> performing its I/O. job->bs isn't used by the mirroring code any more
> afterwards.
>
> Signed-off-by: Kevin Wolf
> ---
> block/mirror.c | 74
> -
On Tue, 05/24 19:47, Laszlo Ersek wrote:
> Which I think satisfies (a->mr == b->mr), but falsifies (a->romd_mode ==
> b->romd_mode).
>
> In effect, the patch seems to allow merging and equality between FlatRange
> objects when they only differ in romd_mode, and that's wrong.
>
> Given that the co
Signed-off-by: Emilio G. Cota
---
tests/.gitignore | 1 +
tests/Makefile | 5 -
tests/test-qht-par.c | 56
3 files changed, 61 insertions(+), 1 deletion(-)
create mode 100644 tests/test-qht-par.c
diff --git a/tests/.gitignore
v5: https://lists.gnu.org/archive/html/qemu-devel/2016-05/msg02366.html
v6 applies cleanly on top of tcg-next (8b1fe3f4 "cpu-exec:
Clean up 'interrupt_request' reloading", tagged "pull-tcg-20160512").
Changes from v5, mostly from Sergey's review:
- processor.h: use #ifdef #elif throughout the fi
This serves as a performance benchmark as well as a stress test
for QHT. We can tweak quite a number of things, including the
number of resize threads and how frequently resizes are triggered.
A performance comparison of QHT vs CLHT[1] and ck_hs[2] using
this same benchmark program can be found he
Reviewed-by: Richard Henderson
Reviewed-by: Alex Bennée
Signed-off-by: Emilio G. Cota
---
include/qemu/compiler.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/qemu/compiler.h b/include/qemu/compiler.h
index 8f1cc7b..b64f899 100644
--- a/include/qemu/compiler.h
+++ b/include/qem
Having a fixed-size hash table for keeping track of all translation blocks
is suboptimal: some workloads are just too big or too small to get maximum
performance from the hash table. The MRU promotion policy helps improve
performance when the hash table is a little undersized, but it cannot
make up
This is a fast, scalable chained hash table with optional auto-resizing,
allowing
reads that are concurrent with reads, and reads/writes that are concurrent
with writes to separate buckets.
A hash table with these features will be necessary for the scalability
of the ongoing MTTCG work; before th
Reviewed-by: Alex Bennée
Reviewed-by: Richard Henderson
Signed-off-by: Emilio G. Cota
---
tests/.gitignore | 1 +
tests/Makefile | 6 ++-
tests/test-qht.c | 159 +++
3 files changed, 165 insertions(+), 1 deletion(-)
create mode 100644 t
Reviewed-by: Richard Henderson
Signed-off-by: Emilio G. Cota
---
tests/.gitignore | 1 +
tests/Makefile | 6 +-
tests/test-qdist.c | 369 +
3 files changed, 375 insertions(+), 1 deletion(-)
create mode 100644 tests/test-qdist.c
diff
Sometimes it is useful to have a quick histogram to represent a certain
distribution -- for example, when investigating a performance regression
in a hash table due to inadequate hashing.
The appended allows us to easily represent a distribution using Unicode
characters. Further, the data structur
For some workloads such as arm bootup, tb_phys_hash is performance-critical.
The is due to the high frequency of accesses to the hash table, originated
by (frequent) TLB flushes that wipe out the cpu-private tb_jmp_cache's.
More info:
https://lists.nongnu.org/archive/html/qemu-devel/2016-03/msg05
This option is unused; besides, it bloats the struct when not needed.
Let's just let writers define their own locks elsewhere.
Reviewed-by: Alex Bennée
Reviewed-by: Richard Henderson
Signed-off-by: Emilio G. Cota
---
cpus.c | 2 +-
include/qemu/seqlock.h | 10 +-
2 fil
From: Guillaume Delbergue
Signed-off-by: Guillaume Delbergue
[Rewritten. - Paolo]
Signed-off-by: Paolo Bonzini
[Emilio's additions: use TAS instead of atomic_xchg; emit acquire/release
barriers; return bool from trylock; call cpu_relax() while spinning;
optimize for uncontended locks by acqui
Examples:
- Good hashing, i.e. tb_hash_func5(phys_pc, pc, flags):
TB count715135/2684354
[...]
TB hash buckets 388775/524288 (74.15% head buckets used)
TB hash occupancy 33.04% avg chain occ. Histogram: [0,10)%|▆ █
▅▁▃▁▁|[90,100]%
TB hash avg chain 1.017 buckets. Histogram: 1
It is a more appropriate name, now that the mutex embedded
in the seqlock is gone.
Reviewed-by: Alex Bennée
Reviewed-by: Richard Henderson
Signed-off-by: Emilio G. Cota
---
cpus.c | 28 ++--
include/qemu/seqlock.h | 4 ++--
2 files changed, 16 insertion
Taken from the linux kernel.
Reviewed-by: Richard Henderson
Reviewed-by: Alex Bennée
Signed-off-by: Emilio G. Cota
---
include/qemu/processor.h | 30 ++
1 file changed, 30 insertions(+)
create mode 100644 include/qemu/processor.h
diff --git a/include/qemu/processo
This will be used by upcoming changes for hashing the tb hash.
Add this into a separate file to include the copyright notice from
xxhash.
Reviewed-by: Richard Henderson
Signed-off-by: Emilio G. Cota
---
include/exec/tb-hash-xx.h | 94 +++
1 file chan
So I needn't disable access to the device between being notified that
the error occurred and being notified to resume.
This will make code simpler.
Am I right?
On 2016/5/24 18:49, Michael S. Tsirkin wrote:
On Tue, Apr 26, 2016 at 08:48:15AM -0600, Alex Williamson wrote:
I think that means tha
> > > > > > > This is grossly inefficient if you only requested a single page.
> > > > > > > And it's also allocating memory very aggressively without
> > > > > > > ever telling the host what is going on.
> > > > > >
> > > > > > If only requested a single page, there is no need to send the
> > > >
> On Tue, May 24, 2016 at 02:36:08PM +, Li, Liang Z wrote:
> > > > > > > This can be pre-initialized, correct?
> > > > > >
> > > > > > pre-initialized? I am not quite understand your mean.
> > > > >
> > > > > I think you can maintain sg as part of device state and init sg
> > > > > with the
> >
On 05/24/2016 07:47 AM, Kevin Wolf wrote:
> This changes the streaming block job to use the job's BlockBackend for
> performing the COR reads. job->bs isn't used by the streaming code any
> more afterwards.
>
> Signed-off-by: Kevin Wolf
> ---
> block/io.c| 9 -
> block/strea
On 05/24/2016 07:47 AM, Kevin Wolf wrote:
> Also add trace points now that the function can be directly called.
>
> Signed-off-by: Kevin Wolf
> ---
> block/block-backend.c | 21 ++---
> include/sysemu/block-backend.h | 6 ++
> trace-events | 4 +++
On 05/24/2016 07:47 AM, Kevin Wolf wrote:
> This adds a new BlockBackend field to the BlockJob struct, which
> coexists with the BlockDriverState while converting the individual jobs.
>
> When creating a block job, a new BlockBackend is created on top of the
> given BlockDriverState, and it is des
On Wed, May 25, 2016 at 01:17:21 +0300, Sergey Fedorov wrote:
> >> With this implementation we could:
> >> (1) get rid of qht_map::stale
> >> (2) don't waste cycles waiting for resize to complete
> > I'll include this in v6.
>
> How is it by perf?
Not much of a difference, since resize is a slo
On 05/24/2016 05:59 AM, Alberto Garcia wrote:
> Signed-off-by: Alberto Garcia
> ---
> docs/throttle.txt | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
Reviewed-by: Eric Blake
>
> diff --git a/docs/throttle.txt b/docs/throttle.txt
> index 06ed9b3..b4431f6 100644
> --- a/docs/throt
On 05/24/2016 02:40 AM, Peter Lieven wrote:
> until now the allocation map was used only as a hint if a cluster
> is allocated or not. If a block was not allocated (or Qemu had
> no info about the allocation status) a get_block_status call was
> issued to check the allocation status and possibly av
On 23/05/16 20:36, Mark Cave-Ayland wrote:
> On 23/05/16 13:54, Paolo Bonzini wrote:
>
>> scsi-block uses the block layer for reads and writes in order to avoid
>> allocating bounce buffers as big as the transferred data. We know how
>> to split a large transfer to multiple reads and writes, and
On 05/24/2016 07:47 AM, Kevin Wolf wrote:
> So far, bdrv_close_all() first removed all root BlockDriverStates of
> BlockBackends and monitor owned BDSes, and then assumed that the
> remaining BDSes must be related to jobs and cancelled these jobs.
>
> This order doesn't work that well any more whe
Commit 983a1600 changed the semantics of blk_write_zeroes() to
be byte-based rather than sector-based, but did not change the
name, which is an open invitation for other code to misuse the
function. Renaming to pwrite_zeroes() makes it more in line
with other byte-based interfaces, and will help m
Another step on our continuing quest to switch to byte-based
interfaces.
Signed-off-by: Eric Blake
---
block/raw_bsd.c | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/block/raw_bsd.c b/block/raw_bsd.c
index d9adf90..b1d5237 100644
--- a/block/raw_bsd.c
+++ b/block
Another step on our continuing quest to switch to byte-based
interfaces.
There are still opportunities to optimize the qcow2 handling
of zero clusters. For example, if the backing file only has
non-zero data in the portion about to be overwritten, then
we could widen the request and make the enti
Another step towards removing sector-based interfaces: convert
the maximum write and minimum alignment values from sectorss to
bytes. Alignment is changed to 'int', since it makes no sense
to have an alignment larger than the maximum write. Add an
assert that no one was trying to use sectors to g
Another step on our continuing quest to switch to byte-based
interfaces.
Signed-off-by: Eric Blake
---
block/gluster.c | 15 +++
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/block/gluster.c b/block/gluster.c
index a8aaacf..15aff4b 100644
--- a/block/gluster.c
+++ b/b
Another step on our continuing quest to switch to byte-based
interfaces.
Signed-off-by: Eric Blake
---
block/vmdk.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/block/vmdk.c b/block/vmdk.c
index 8494d63..284d7a0 100644
--- a/block/vmdk.c
+++ b/block/vmdk.c
@@
Rename to bdrv_pwrite_zeroes() to let the compiler ensure we
cater to the updated semantics. Do the same for
bdrv_aio_write_zeroes() and bdrv_co_write_zeroes(). For now,
we still require sector alignment in the callers, via assertions.
Signed-off-by: Eric Blake
---
include/block/block.h | 16
Another step on our continuing quest to switch to byte-based
interfaces.
Signed-off-by: Eric Blake
---
block/raw-posix.c | 37 +++--
trace-events | 2 +-
2 files changed, 16 insertions(+), 23 deletions(-)
diff --git a/block/raw-posix.c b/block/raw-posix.c
i
Another step on our continuing quest to switch to byte-based
interfaces.
Kill an abuse of the comma operator while at it (fortunately,
the semantics were still right).
Signed-off-by: Eric Blake
---
block/qed.c | 25 +
1 file changed, 13 insertions(+), 12 deletions(-)
di
Another step on our continuing quest to switch to byte-based
interfaces.
Signed-off-by: Eric Blake
---
block/blkreplay.c | 10 --
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/block/blkreplay.c b/block/blkreplay.c
index 1a721ad..525c2d5 100755
--- a/block/blkreplay.c
+++
Update bdrv_co_do_write_zeroes() to be byte-based, and select
between the new byte-based bdrv_co_pwrite_zeroes() or the old
bdrv_co_write_zeroes(). The next patches will convert drivers,
then remove the old interface.
Signed-off-by: Eric Blake
---
include/block/block_int.h | 4 ++-
block/io.c
Now that all drivers have been converted to a byte interface,
we no longer need a sector interface.
Signed-off-by: Eric Blake
---
include/block/block_int.h | 2 --
block/io.c| 15 ++-
2 files changed, 2 insertions(+), 15 deletions(-)
diff --git a/include/block/block
Another step on our continuing quest to switch to byte-based
interfaces.
As this is the first byte-based iscsi interface, convert
is_request_lun_aligned() into two versions, one for sectors
and one for bytes.
Signed-off-by: Eric Blake
---
block/iscsi.c | 53 +++--
Kevin pointed out that my recent change to byte-based instead
of sector-based blk_write_zeroes() (commit 983a1600) makes life
harder as long as bdrv_write_zeroes is still sector-based, and
where the compiler doesn't flag any change in parameter types.
Complete the conversion, by renaming things (so
On 25/05/16 01:07, Emilio G. Cota wrote:
> On Mon, May 23, 2016 at 23:28:27 +0300, Sergey Fedorov wrote:
>> What if we turn qht::lock into a mutex and change the function as follows:
>>
>> static inline
>> struct qht_bucket *qht_bucket_lock__no_stale(struct qht *ht,
>> uint32_t hash,
>>
On Mon, May 23, 2016 at 23:28:27 +0300, Sergey Fedorov wrote:
> What if we turn qht::lock into a mutex and change the function as follows:
>
> static inline
> struct qht_bucket *qht_bucket_lock__no_stale(struct qht *ht,
> uint32_t hash,
>
On 05/13/2016 05:37 PM, Alistair Francis wrote:
+if (elf_machine < 1) {
+/* The caller didn't specify and ARCH, we can figure it out */
Spotted a comment typo: s/and/an/
+lseek(fd, 0x12, SEEK_SET);
+if (read(fd, &e_machine, sizeof(e_machine)) != sizeof(e_machine)
On 05/04/16 18:32, Alex Bennée wrote:
> diff --git a/cpu-exec.c b/cpu-exec.c
> index bd50fef..f558508 100644
> --- a/cpu-exec.c
> +++ b/cpu-exec.c
> @@ -28,6 +28,7 @@
> #include "qemu/rcu.h"
> #include "exec/tb-hash.h"
> #include "exec/log.h"
> +#include "qemu/main-loop.h"
> #if defined(TARGET_
On Tue, May 24, 2016 at 23:09:47 +0300, Sergey Fedorov wrote:
> On 24/05/16 23:06, Emilio G. Cota wrote:
> > For correctness, smp_read_barrier_depends() is only required to
> > emit a barrier on Sparc hosts. However, we are currently emitting
> > a consume fence unconditionally.
> >
> > Fix it by k
Currently we emit a consume-load in atomic_rcu_read. This is
overkill for non-Sparc hosts, and is only useful to make
things easier for Thread Sanitizer, which as far as I understand
works best without explicit fences.
The appended leaves the consume-load in atomic_rcu_read when
compiling with Thr
On 24/05/16 23:06, Emilio G. Cota wrote:
> For correctness, smp_read_barrier_depends() is only required to
> emit a barrier on Sparc hosts. However, we are currently emitting
> a consume fence unconditionally.
>
> Fix it by keeping the consume fence if we're compiling with Thread
> Sanitizer, since
Recently Linux did a mass conversion of its atomic_read/set calls
so that they at least are READ/WRITE_ONCE. See Linux's commit
62e8a325 ("atomic, arch: Audit atomic_{read,set}()"). It seems though
that their documentation hasn't been updated to reflect this.
The appended updates our documentation
For correctness, smp_read_barrier_depends() is only required to
emit a barrier on Sparc hosts. However, we are currently emitting
a consume fence unconditionally.
Fix it by keeping the consume fence if we're compiling with Thread
Sanitizer, since this might help prevent false warnings. Otherwise,
v1: https://lists.gnu.org/archive/html/qemu-devel/2016-05/msg03661.html
Patch 1 hasn't changed from v1 (where it was patch 2 though).
Patches 2 and 3 fix a not-so-small-after-all RCU performance regression
we introduced when transitioning to __atomic primitives. I got
an arm64 machine to test tod
On 24/05/16 22:56, Emilio G. Cota wrote:
> On Tue, May 24, 2016 at 09:08:01 +0200, Paolo Bonzini wrote:
>> On 23/05/2016 19:09, Emilio G. Cota wrote:
>>> PS. And really equating smp_wmb/rmb to release/acquire as we have under
>>> #ifdef __ATOMIC is hard to justify, other than to please tsan.
>> Tha
VFIO Type1 IOMMU driver is designed for the devices which are IOMMU
capable. Mediated device only uses IOMMU TYPE1 API, the underlying
hardware can be managed by an IOMMU domain.
This change exports functions to pin and unpin pages for mediated devices.
It maintains data of pinned pages for mediat
1 - 100 of 313 matches
Mail list logo
