this title confused me
i think you want to express "only compare the packets that has the same sequence
number"
and i think this optimization can not reduce this checkpoint optimization.
Thanks
On 07/07/2017 01:43 PM, Zhang Chen wrote:
If primary packet's sequence number not same with secon
On 12/04/2018 09:06 AM, Michael S. Tsirkin wrote:
On Mon, Dec 03, 2018 at 10:48:51PM +0800, Li Zhijian wrote:
It provides setup_data struct and header fields
CC: Michael S. Tsirkin
Signed-off-by: Li Zhijian
Sorry I was unclear when I said "import it from Linux".
Sorry, it&
lude/asm/unistd_x32.h" "$output/linux-headers/asm-x86/"
cp "$tmpdir/include/asm/unistd_64.h" "$output/linux-headers/asm-x86/"
cp_portable "$tmpdir/include/asm/kvm_para.h"
"$output/include/standard-headers/asm-$arch"
+
On 12/05/2018 01:40 AM, Philippe Mathieu-Daudé wrote:
Hi Li,
On 3/12/18 15:48, Li Zhijian wrote:
Some address/memory APIs have different type between
'hwaddr/target_ulong addr' and 'int len'. It is very unsafety, espcially
I'm not native English speaker, but I t
On 12/05/2018 11:33 PM, Michael S. Tsirkin wrote:
On Wed, Dec 05, 2018 at 06:28:11PM +0800, Li Zhijian wrote:
Hi Michael
I cooked a draft with cp_portable to import bootparam.h, could you have a look.
diff --git a/scripts/update-linux-headers.sh b/scripts/update-linux-headers.sh
index
enough if
this bit is set.
CC: Paolo Bonzini
CC: Richard Henderson
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
Signed-off-by: Li Zhijian
---
V3: correct grammar and check XLF_CAN_BE_LOADED_ABOVE_4G first (Michael S.
Tsirkin)
Signed-off-by: Li Zhijian
--
it's from v4.20-rc5.
CC: Michael S. Tsirkin
Signed-off-by: Li Zhijian
---
V4: use scirpt to import bootparam.h (Michael S. Tsirkin)
V3: new patch
Signed-off-by: Li Zhijian
---
hw/i386/pc.c | 8 +--
include/standard-headers/asm-x86/bootparam.h
eader bootparam.h (Michael S. Tsirkin)
V2: add 2 patches(3/5, 4/5) to fix potential loading issue.
Li Zhijian (4):
unify len and addr type for memory/address APIs
refactor load_image_size
i386: import & use bootparam.h
i386: allow to load initrd below 4G for recent linux
Don't expect read(2) can always read as many as it's told.
Signed-off-by: Li Zhijian
Reviewed-by: Richard Henderson
---
V4: add reviewed-by tag
---
hw/core/loader.c | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/hw/core/loader.c b/hw/core/load
ry_read(uint32_t len)
-> dma_memory_rw(uint32_t len)
-> dma_memory_rw_relaxed(uint32_t len)
-> address_space_rw(int len) # len overflow
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
CC: Peter Maydell
Signed-off-by: Li Zhijian
Reviewed-by: Peter Maydell
Reviewed-b
On 12/06/2018 04:37 AM, Marc-André Lureau wrote:
COLO uses a worker context (iothread) to drive the chardev. All
backends are not able to switch the context, let's report an error in
this case.
Signed-off-by: Marc-André Lureau
Reviewed-by: Li Zhijian
---
net/colo-compare.
w_relaxed(uint32_t len)
-> address_space_rw(int len) # len overflow
-> address_space_read_full(int len)
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
CC: Philip Li
Signed-off-by: Li Zhijian
---
exec.c| 42 +---
allow load_image to load >= 2G file
CC: Philip Li
Signed-off-by: Li Zhijian
---
hw/core/loader.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/hw/core/loader.c b/hw/core/loader.c
index aa0b3fc..8fbc4bd 100644
--- a/hw/core/loader.c
+++ b/hw/core/loader.c
@@ -77,7 +7
: Li Zhijian
---
hw/i386/pc.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index cd5029c..e1b910f 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
@@ -913,6 +913,12 @@ static void load_linux(PCMachineState *pcms,
/* highest address for loading the initrd
$ make # all errors gone
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
Li Zhijian (3):
i386: set initrd_max to 4G - 1 to allow up to 4G initrd
change size type from int to int64_t on load_image()
change int len to uin32_t len
exec.c| 42 +++
On 11/08/2018 07:06 PM, Peter Maydell wrote:
On 8 November 2018 at 10:59, Li Zhijian wrote:
x86/x86_64 has alredy supported 4G initrd.
linux/arch/x86/boot/header.S:
# (Header version 0x0203 or later) the highest safe address for the contents
# of an initrd. The current kernel allows up
On 11/08/2018 07:04 PM, Peter Maydell wrote:
On 8 November 2018 at 10:59, Li Zhijian wrote:
allow load_image to load >= 2G file
CC: Philip Li
Signed-off-by: Li Zhijian
---
hw/core/loader.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/hw/core/loader.c b/hw/c
- If 1, kernel/boot_params/cmdline/ramdisk can be above 4G.
629
maybe we can reuse this field and append a new Bit 5 XLF_INITRD_SIZE_4G or such
thanks
Zhijian
On 11/9/2018 6:04 PM, Juergen Gross wrote:
On 09/11/2018 10:57, Li Zhijian wrote:
On 11/9/2018 3:20 PM, Ingo Molnar wrote
.
bootloader side:
if protocol >= 2.15
if XLF_INITRD_LOAD_BELOW_4G
support ~4G initrd
fi
else if protocol >=2.12
if XLF_CAN_BE_LOADED_ABOVE_4G
support ~4G initrd
fi
fi
thanks
Zhijian
On 11/9/2018 6:04 PM, Juergen Gross wrote:
On 09/11/2018 10:57, Li Zhijian wrote:
O
On 11/9/2018 3:20 PM, Ingo Molnar wrote:
* Li Zhijian wrote:
If the kernel initrd creation process creates an initrd which
is larger than 2GB and also claims that it can't be placed
with any part of it above 2GB, then that sounds like a bug
in the initrd creation process...
Exactly, i
This patch allow load_iamge to load >=2G file
Signed-off-by: Li Zhijian
---
hw/core/loader.c| 5 +++--
include/hw/loader.h | 2 +-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/hw/core/loader.c b/hw/core/loader.c
index aa0b3fc..0d53229 100644
--- a/hw/core/loader.c
+++ b
this bit is set.
CC: Paolo Bonzini
CC: Richard Henderson
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
Signed-off-by: Li Zhijian
---
hw/i386/pc.c | 12
1 file changed, 12 insertions(+)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index f095725..5e2f
-> dma_memory_rw(uint32_t len)
-> dma_memory_rw_relaxed(uint32_t len)
-> address_space_rw(int len) # len overflow
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
Signed-off-by: Li Zhijian
---
exec.c| 49
oot@haswell-OptiPlex-9020:/home/lizj# du -sh large.cgz
2.5Glarge.cgz
this patch changes the caller side that use this function to calculate
size of initrd file as well.
Signed-off-by: Li Zhijian
---
v3: hide version changelog
v2: update error message and int64_t printing format
hw/alpha/dp26
This patch allow load_iamge to load >=2G file
Signed-off-by: Li Zhijian
---
hw/core/loader.c| 5 +++--
include/hw/loader.h | 2 +-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/hw/core/loader.c b/hw/core/loader.c
index aa0b3fc..0d53229 100644
--- a/hw/core/loader.c
+++ b
-> dma_memory_rw(uint32_t len)
-> dma_memory_rw_relaxed(uint32_t len)
-> address_space_rw(int len) # len overflow
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
Signed-off-by: Li Zhijian
---
exec.c| 49
len = len > INT32_MAX ? INT32_MAX : len;
+
/* If the access is not a read access, it will be a skip access,
* tested before.
*/
changes:
V2: add 2 patches(3/5, 4/5) to fix potential loading issue.
Li Zhijian (5):
unify len and addr type for memory/addre
Don't expect read(2) can always read as many as it's told.
Signed-off-by: Li Zhijian
---
hw/core/loader.c | 24 ++--
1 file changed, 14 insertions(+), 10 deletions(-)
diff --git a/hw/core/loader.c b/hw/core/loader.c
index 0d53229..5f891e2 100644
--- a/hw/core/load
this bit is set.
CC: Paolo Bonzini
CC: Richard Henderson
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
Signed-off-by: Li Zhijian
---
hw/i386/pc.c | 12
1 file changed, 12 insertions(+)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 2ffe6fb..6d4b
CC: Paolo Bonzini
CC: Richard Henderson
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
Signed-off-by: Li Zhijian
---
hw/i386/pc.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index f095725..2ffe6fb 10064
On 11/30/2018 10:53 PM, Michael S. Tsirkin wrote:
On Wed, Nov 21, 2018 at 10:06:06AM +0800, Li Zhijian wrote:
a new field xloadflags was added to recent x86 linux, and BIT 1:
XLF_CAN_BE_LOADED_ABOVE_4G is used to tell bootload that where initrd can be
loaded saftly.
safely
Current QEMU
On 11/30/2018 11:14 PM, Peter Maydell wrote:
I'm just about to send out a patchset which removes all the
load_image() callers; that would make patches 2 and 4 in this
set unnecessary.
Got it, i will remove them at next version basing on your patch set.
Thanks
Zhijian
On 11/30/2018 09:40 PM, Peter Maydell wrote:
On Wed, 21 Nov 2018 at 02:07, Li Zhijian wrote:
Some address/memory APIs have different type between 'hwaddr addr' and
'int len'. It is very unsafety, espcially some APIs will be passed a non-int
len by caller which might ca
It provides setup_data struct and header fields
CC: Michael S. Tsirkin
Signed-off-by: Li Zhijian
---
V3: new patch
---
hw/i386/pc.c | 15 +--
1 file changed, 1 insertion(+), 14 deletions(-)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
index 067d23a..8db7417 100644
--- a/hw/i386/pc.c
enough if
this bit is set.
CC: Paolo Bonzini
CC: Richard Henderson
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
Signed-off-by: Li Zhijian
---
V3: correct grammar and check XLF_CAN_BE_LOADED_ABOVE_4G first (Michael S.
Tsirkin)
Signed-off-by: Li Zhijian
--
code basing on http://patchwork.ozlabs.org/cover/1005990 and
https://patchew.org/QEMU/20181122133507.30950-1-peter.mayd...@linaro.org
- add new patch 3/4 to import header bootparam.h (Michael S. Tsirkin)
V2: add 2 patches(3/5, 4/5) to fix potential loading issue.
Li Zhijian (4):
unify len and
Don't expect read(2) can always read as many as it's told.
Signed-off-by: Li Zhijian
---
hw/core/loader.c | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/hw/core/loader.c b/hw/core/loader.c
index fa41842..9cbceab 100644
--- a/hw/core/loader.c
+++
ry_read(uint32_t len)
-> dma_memory_rw(uint32_t len)
-> dma_memory_rw_relaxed(uint32_t len)
-> address_space_rw(int len) # len overflow
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
CC: Peter Maydell
Signed-off-by: Li Zhijian
---
V3: use the same type betwe
ping
On 12/6/18 10:32, Li Zhijian wrote:
Long long ago, linux kernel has supported up to 4G initrd, but it's header
still hard code to allow loading initrd below 2G only.
cutting from arch/x86/head.S:
# (Header version 0x0203 or later) the highest safe address for the contents
#
The reasons for not having a /dev/kvm can vary but you haven't said what
your host architecture is and what OS you are running so it's hard to
offer advice here.
--
Alex Bennée
--
Best regards.
Li Zhijian (8528)
On 12/22/18 00:12, Michael S. Tsirkin wrote:
On Thu, Dec 06, 2018 at 10:32:11AM +0800, Li Zhijian wrote:
Don't expect read(2) can always read as many as it's told.
Signed-off-by: Li Zhijian
Reviewed-by: Richard Henderson
This is more a theoretical bugfix than a refactoring ri
ry_read(uint32_t len)
-> dma_memory_rw(uint32_t len)
-> dma_memory_rw_relaxed(uint32_t len)
-> address_space_rw(int len) # len overflow
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
CC: Peter Maydell
CC: Stefano Garzarella
Signed-off-by: Li Zhijian
Reviewed-b
Don't expect read(2) can always read as many as it's told.
CC: Richard Henderson
CC: Stefano Garzarella
Signed-off-by: Li Zhijian
Reviewed-by: Richard Henderson
Reviewed-by: Stefano Garzarella
---
V5: update subject and add reviewed-by tag (Stefano Garzarella)
V4: add revie
ocol (specifically the ext_ramdisk_image field).
Therefore here just limit initrd_max to UINT32_MAX simply as well to
allow initrd to be loaded below 4 GB.
CC: Paolo Bonzini
CC: Richard Henderson
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
Signed-off-by: Li
it's from v4.20-rc5.
CC: Michael S. Tsirkin
Signed-off-by: Li Zhijian
Reviewed-by: Michael S. Tsirkin
---
V5: add reviewed-by tag
V4: use scirpt to import bootparam.h (Michael S. Tsirkin)
V3: new patch
---
hw/i386/pc.c | 8 +--
include/standard-header
ot;
CC: Marcel Apfelbaum
CC: Stefano Garzarella
CC: Peter Crosthwaite
CC: Peter Maydell
Li Zhijian (4):
unify len and addr type for memory/address APIs
hw/core/loader.c: Read as long as possible in load_image_size()
i386: import & use bootparam.h
i386: allow to load initrd below
Hi Stefano
On 1/11/19 17:48, Stefano Garzarella wrote:
Hi Li,
On Fri, Jan 11, 2019 at 10:06 AM Li Zhijian wrote:
+# unnecessary import of several video/ist/etc headers
+sed -e '/__ASSEMBLY__/,/__ASSEMBLY__/d' $tmpdir/include/asm/bootparam.h
> $tmpdi
Hi Eduardo
On 1/15/19 01:53, Eduardo Habkost wrote:
+if (protocol >= 0x20c &&
+lduw_p(header+0x236) & XLF_CAN_BE_LOADED_ABOVE_4G) {
+/*
+ * Linux has supported initrd up to 4 GB for a very long time (2007,
+ * long before XLF_CAN_BE_LOADED_ABOVE_4G which was
Hi Michael, Eduardo
On 1/15/19 09:46, Michael S. Tsirkin wrote:
On Tue, Jan 15, 2019 at 09:35:09AM +0800, Li Zhijian wrote:
Hi Eduardo
On 1/15/19 01:53, Eduardo Habkost wrote:
+if (protocol >= 0x20c &&
+lduw_p(header+0x236) & XLF_CAN_BE_
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
Signed-off-by: Li Zhijian
---
V6: update comments
V5: udpate comments and changelog
V3: correct grammar and check XLF_CAN_BE_LOADED_ABOVE_4G first (Michael S.
Tsirkin)
Signed-off-by: Li Zhijian
---
hw/i386/pc.c |
kin)
V2: add 2 patches(3/5, 4/5) to fix potential loading issue.
CC: Paolo Bonzini
CC: Richard Henderson
CC: Eduardo Habkost
CC: "Michael S. Tsirkin"
CC: Marcel Apfelbaum
CC: Stefano Garzarella
CC: Peter Crosthwaite
CC: Peter Maydell
Li Zhijian (4):
unify len and addr type
it's from v4.20-rc5.
CC: Stefano Garzarella
CC: Michael S. Tsirkin
Signed-off-by: Li Zhijian
Reviewed-by: Michael S. Tsirkin
Reviewed-by: Stefano Garzarella
---
V6: Fix line over 80 characters && use double quates for all pathes (Stefano
Garzarella)
V5: add reviewed-by tag
V4:
ry_read(uint32_t len)
-> dma_memory_rw(uint32_t len)
-> dma_memory_rw_relaxed(uint32_t len)
-> address_space_rw(int len) # len overflow
CC: Paolo Bonzini
CC: Peter Crosthwaite
CC: Richard Henderson
CC: Peter Maydell
CC: Stefano Garzarella
Signed-off-by: Li Zhijian
Reviewed-b
Don't expect read(2) can always read as many as it's told.
CC: Richard Henderson
CC: Stefano Garzarella
Signed-off-by: Li Zhijian
Reviewed-by: Richard Henderson
Reviewed-by: Stefano Garzarella
---
V5: update subject and add reviewed-by tag (Stefano Garzarella)
V4: add revie
timeout is updated to 5mins since we need more time to load a
large initrd to the guest
CC: Wainer dos Santos Moschetta
CC: Caio Carrara
CC: Cleber Rosa
CC: Eduardo Habkost
CC: Philippe Mathieu-Daudé
Signed-off-by: Li Zhijian
---
tests/acceptance/linux_initrd.py | 37
linux-3.6 kernel shipped by Fedora-19 cannot support xldflags so that it
cannot support loading more than 2GiB initrd
CC: Wainer dos Santos Moschetta
CC: Caio Carrara
CC: Cleber Rosa
CC: Eduardo Habkost
CC: Philippe Mathieu-Daudé
Signed-off-by: Li Zhijian
---
this patch is basing on early
Sorry, there are some typos, please ignore this version
i will correct in V2
On 1/18/19 18:02, Li Zhijian wrote:
linux-3.6 kernel shipped by Fedora-19 cannot support xldflags so that it
s/Fedora-19/Fedora-18
cannot support loading more than 2GiB initrd
CC: Wainer dos Santos Moschetta
CC
Sorry, there are some typos, please ignore this version
i will correct in V2
s/v4.19/v4.16 on subject.
On 1/18/19 18:02, Li Zhijian wrote:
timeout is updated to 5mins since we need more time to load a
large initrd to the guest
CC: Wainer dos Santos Moschetta
CC: Caio Carrara
CC: Cleber
On 12/28/2018 4:31 AM, Eduardo Habkost wrote:
On Fri, Dec 21, 2018 at 11:10:30AM -0500, Michael S. Tsirkin wrote:
On Thu, Dec 06, 2018 at 10:32:13AM +0800, Li Zhijian wrote:
/* highest address for loading the initrd */
-if (protocol >= 0x203) {
+if (protocol >=
On 1/7/19 18:04, Stefano Garzarella wrote:
As Philippe already suggested,
s/espcially/especially
Hi Stefano，
thanks a lot, will update it at next version.
Thanks
On 1/7/19 18:33, Stefano Garzarella wrote:
On Mon, Dec 24, 2018 at 3:16 AM Li Zhijian wrote:
On 12/22/18 00:12, Michael S. Tsirkin wrote:
On Thu, Dec 06, 2018 at 10:32:11AM +0800, Li Zhijian wrote:
Don't expect read(2) can always read as many as it's told.
Signed-off-by:
Hi Stefano,
On 1/5/19 00:41, Stefano Garzarella wrote:
+# Remove everything except the macros from bootparam.h avoiding the
+# unnecessary import of several video/ist/etc headers
+sed -e '/__ASSEMBLY__/,/__ASSEMBLY__/d' $tmpdir/include/asm/bootparam.h
> $tmpdir/bootpara
On 1/7/19 20:11, Stefano Garzarella wrote:
Hi,
On Thu, Dec 27, 2018 at 9:32 PM Eduardo Habkost wrote:
On Fri, Dec 21, 2018 at 11:10:30AM -0500, Michael S. Tsirkin wrote:
On Thu, Dec 06, 2018 at 10:32:13AM +0800, Li Zhijian wrote:
a new field xloadflags was added to recent x86 linux, and BIT
oot@haswell-OptiPlex-9020:/home/lizj# du -sh large.cgz
2.5Glarge.cgz
this patch changes the caller side that use this function to calculate
size of initrd file as well.
Signed-off-by: Li Zhijian
---
hw/alpha/dp264.c| 2 +-
hw/core/loader.c| 5 +++--
hw/hppa/machine.c
oot@haswell-OptiPlex-9020:/home/lizj# du -sh large.cgz
2.5Glarge.cgz
this patch changes the caller side that use this function to calculate
size of initrd file as well.
v2: update error message and int64_t printing format
Signed-off-by: Li Zhijian
---
hw/alpha/dp264.c| 3 ++-
hw/c
Signed-off-by: Li Zhijian
---
migration/savevm.c | 53 +++--
1 file changed, 35 insertions(+), 18 deletions(-)
diff --git a/migration/savevm.c b/migration/savevm.c
index 33a2911..09d8e99 100644
--- a/migration/savevm.c
+++ b/migration/savevm.c
migration, source keep in status "running".
Signed-off-by: Li Zhijian
---
migration/savevm.c | 14 +++---
1 file changed, 11 insertions(+), 3 deletions(-)
diff --git a/migration/savevm.c b/migration/savevm.c
index 09d8e99..0c07671 100644
--- a/migration/savevm.c
+++ b/migratio
On 09/30/2016 02:15 PM, Amit Shah wrote:
Hi,
On (Thu) 29 Sep 2016 [19:06:32], Li Zhijian wrote:
Priviously, if the source and distination have different devices, source could
goto
the status "paused (postmigrate)", and the distination will exit that means no
qemu
is alive.
this bug occurred in COLO frame related codes. This part of codes wrote
by Li zhijian and Zhang hailiang and currently maintained by Zhang hailiang.
So, I add them to this thread.
CC Zhijian and Hailiang:
Any idea or comments about this bug?
One clue is the memory of SVM not is same with PVM.
we
Just noticed that's a little old, you may need to rebase it
Thanks
On 03/23/2018 11:51 AM, Li Zhijian wrote:
On 03/21/2018 02:04 PM, Zhang Chen wrote:
Hi Suiheng,
I made a new guest image and retest it, and got the same bug from latest branch.
I found that after the COLO checkpoint
Signed-off-by: Li Zhijian
---
.gitignore | 1 +
1 file changed, 1 insertion(+)
diff --git a/.gitignore b/.gitignore
index 2992d15931..28729241f1 100644
--- a/.gitignore
+++ b/.gitignore
@@ -57,6 +57,7 @@
/qemu-ga
/qemu-bridge-helper
/qemu-keymap
+/virtiofsd
/qemu-monitor.texi
/qemu-monitor
we go here either (!(*iov)[i].iov_base) or (len != l), so we need to consider
to unmap the 'i'th item as well when the 'i'th item is not nil
Signed-off-by: Li Zhijian
---
hw/display/virtio-gpu.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/displ
On 8/21/20 6:07 PM, Gerd Hoffmann wrote:
On Fri, Aug 21, 2020 at 04:49:45PM +0800, Li Zhijian wrote:
we go here either (!(*iov)[i].iov_base) or (len != l), so we need to consider
to unmap the 'i'th item as well when the 'i'th item is not nil
Signed-off-by: Li Zhiji
we go here either (!(*iov)[i].iov_base) or (len != l), so we need to consider
to unmap the 'i'th item as well when the 'i'th item is not nil
Signed-off-by: Li Zhijian
---
v2: address Gerd's comments
---
hw/display/virtio-gpu.c | 3 +++
1 file changed, 3 insertions(+)
the error path after calling qemu_rdma_dest_init() should do rdma cleanup
Signed-off-by: Li Zhijian
Reviewed-by: Dr. David Alan Gilbert
---
migration/rdma.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 41726cc74a8
later.
Signed-off-by: Li Zhijian
---
V2: alloc memory for host_port
---
migration/rdma.c | 32 +---
1 file changed, 21 insertions(+), 11 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 651534e8255..d829d08d076 100644
--- a/migration/rdma.c
+++ b
rdma_freeaddrinfo() is the reverse operation of rdma_getaddrinfo()
Signed-off-by: Li Zhijian
Reviewed-by: Dr. David Alan Gilbert
---
migration/rdma.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/migration/rdma.c b/migration/rdma.c
index 7e7595faabf..651534e8255 100644
--- a/migration
source side always blocks if postcopy is only enabled at source side.
users are not able to cancel this migration in this case.
Let source side have chance to cancel this migration
Signed-off-by: Li Zhijian
---
V2: utilize poll to check cm event
---
migration/rdma.c | 42
Signed-off-by: Li Zhijian
---
migration/rdma.c | 11 ---
1 file changed, 8 insertions(+), 3 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 1cdb4561f32..d90b29a4b51 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -1539,16 +1539,20 @@ static int
_event after rdma_connect");
2499 ERROR(errp, "connecting to destination!");
2500 rdma_ack_cm_event(cm_event); <<<< cause segmentation fault
2501 goto err_rdma_source_connect;
2502 }
Signed-off-by: Li Zhijian
---
migration/rdma.c | 2 +-
1 f
dated/changed only if rdma_get_cm_event() returns 0. So it's okey to
remove the ack in error patch.
Signed-off-by: Li Zhijian
---
V2: remove ack from the error patch (Dave)
---
migration/rdma.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 00eac342
rdma_freeaddrinfo() is the reverse operation of rdma_getaddrinfo()
Signed-off-by: Li Zhijian
---
migration/rdma.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/migration/rdma.c b/migration/rdma.c
index 7e7595faab..651534e825 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
the error path after calling qemu_rdma_dest_init() should do rdma cleanup
Signed-off-by: Li Zhijian
---
migration/rdma.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 41726cc74a..7e7595faab 100644
--- a/migration/rdma.c
enabled postcopy later.
Signed-off-by: Li Zhijian
---
migration/rdma.c | 29 ++---
1 file changed, 18 insertions(+), 11 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 651534e825..3b228c46eb 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -316,6
the error path after calling qemu_rdma_dest_init() should do rdma cleanup
Signed-off-by: Li Zhijian
---
migration/rdma.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 41726cc74a..7e7595faab 100644
--- a/migration/rdma.c
rdma_freeaddrinfo() is the reverse operation of rdma_getaddrinfo()
Signed-off-by: Li Zhijian
---
migration/rdma.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/migration/rdma.c b/migration/rdma.c
index 7e7595faab..651534e825 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
later.
Signed-off-by: Li Zhijian
---
migration/rdma.c | 29 ++---
1 file changed, 18 insertions(+), 11 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 651534e825..3b228c46eb 100644
--- a/migration/rdma.c
+++ b/migration/rdma.c
@@ -316,6 +316,7 @@ typedef
source side always blocks if postcopy is only enabled at source side.
users are not able to cancel this migration in this case.
Here we try to get the cm_event every 100ms tile timeout.
Signed-off-by: Li Zhijian
---
migration/rdma.c | 59
1 file
On 2/5/21 5:52 PM, Claudio Fontana wrote:
Hi LiZhiJian,
just one thing I noticed in your call stack, your $Subject talks about AArch64,
and you end up in arm_v7m_mmu_idx_for_secstate,
which means that ARM_FEATURE_M is detected, so definitely something looks off
when it comes to the feature
And change the default to true so that '-incoming defer' can enable
multifd first.
Signed-off-by: Li Zhijian
---
migration/migration.c | 8
migration/multifd.c | 2 +-
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/migration/migration.c b/migration/migrati
=0x563b4c44bb80) at
../util/qemu-thread-posix.c:541
#5 0x7f7fe9b5b3f9 in ?? ()
#6 0x in ?? ()
It's enough to check migrate_multifd_is_allowed() in multifd cleanup() and
multifd setup() though there are so many other places using
migrate_use_multifd().
Signed-off-by: Li Zh
in some modern HCA hardware/drivers.
For the (2), IB provides advise API to prefetch pages in specific memory
region. It can help driver reduce the page fault on responder(destination)
side during RDMA_WRITE.
CC: marcel.apfelb...@gmail.com
Li Zhijian (2):
migration/rdma: Try to register On
the responder does page fault as little as possible.
Signed-off-by: Li Zhijian
Reviewed-by: Marcel Apfelbaum
---
V2: use IBV_ADVISE_MR_FLAG_FLUSH instead of IB_UVERBS_ADVISE_MR_FLAG_FLUSH
and add Reviewed-by tag. # Marcel
---
migration/rdma.c | 40
]: http://pmem.io/rpma/manpages/v0.9.0/rpma_mr_reg.3
CC: Marcel Apfelbaum
Signed-off-by: Li Zhijian
---
V2: add ODP sanity check and remove goto
---
migration/rdma.c | 73 ++
migration/trace-events | 1 +
2 files changed, 54 insertions(+), 20 deletions
wait Y CQ event(dropped at (7), blocks
forever)
Looks it only happens on soft RoCE rdma device in my a hundred of runs,
a hardward IB device works fine.
Signed-off-by: Li Zhijian
---
This is just a draft to address this problem. One possible approach
could be creating their
The prompt was updated since def835f0da ('hostmem: Don't report pmem attribute
if unsupported')
Signed-off-by: Li Zhijian
---
docs/nvdimm.txt | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/docs/nvdimm.txt b/docs/nvdimm.txt
index 0aae682be3e..71cdbdf554b 10
]: http://pmem.io/rpma/manpages/v0.9.0/rpma_mr_reg.3
Signed-off-by: Li Zhijian
---
migration/rdma.c | 27 ++-
migration/trace-events | 1 +
2 files changed, 19 insertions(+), 9 deletions(-)
diff --git a/migration/rdma.c b/migration/rdma.c
index 5c2d113aa94..8784b5f22a6
some modern HCA hardware/drivers.
For the (2), IB provides advise API to prefetch pages in specific memory
region. It can help driver reduce the page fault on responder(destination)
side during RDMA_WRITE.
Li Zhijian (2):
migration/rdma: Try to register On-Demand Paging memory region
migration
the responder does page fault as little as possible.
Signed-off-by: Li Zhijian
---
migration/rdma.c | 40
migration/trace-events | 1 +
2 files changed, 41 insertions(+)
diff --git a/migration/rdma.c b/migration/rdma.c
index 8784b5f22a6
And change the default to true so that in '-incoming defer' case, user is able
to change multifd capability.
Signed-off-by: Li Zhijian
---
migration/migration.c | 8
migration/multifd.c | 2 +-
2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/migration/mig
101 - 200 of 295 matches
Mail list logo
