[PATCH v4 18/23] migration: Make ram_save_target_page() a pointer

We are going to create a new function for multifd latest in the series. Signed-off-by: Juan Quintela --- migration/ram.c | 12 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/migration/ram.c b/migration/ram.c index e9dcd3ca4e..3536778e19 100644 --- a/migration/ram.c

[PATCH v5 3/5] ppc/pnv: turn 'phb' into a pointer in struct PnvPhb4PecStack

At this moment, stack->phb is the plain PnvPHB4 device itself instead of a pointer to the device. This will present a problem when adding user creatable devices because we can't deal with this struct and the realize() callback from the user creatable device. We can't get rid of this attribute,

Re: [PATCH v3 3/6] migration: ram_release_pages() always receive 1 page as argument

On 1/11/22 13:45, Juan Quintela wrote: > Remove the pages argument. And s/pages/page/ > > Signed-off-by: Juan Quintela > Reviewed-by: Philippe Mathieu-Daudé > Reviewed-by: Peter Xu > > -- Note, you need '---' (3x) to have git tools strip the rest (otherwise this text will be committed). > -

[PATCH v4 17/23] multifd: Rename pages_used to normal_pages

Signed-off-by: Juan Quintela Reviewed-by: Dr. David Alan Gilbert --- migration/multifd.h | 3 ++- migration/multifd.c | 4 ++-- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/migration/multifd.h b/migration/multifd.h index be460f821b..4dda900a0b 100644 --- a/migration/multifd.h

[PATCH v5 0/5] user creatable pnv-phb4 devices

Hi, This version implements Cedric's review suggestions from v4. No drastic design changes were made. Changes from v4: - patches 1,3,5: unchanged - patch 2: * renamed function to pnv_phb4_xscom_realize() * pnv4_phb4_xscom_realize() is now called at the end of phb4_realize() - patch 4: *

[PATCH v4 08/23] multifd: Move iov from pages to params

This will allow us to reduce the number of system calls on the next patch. Signed-off-by: Juan Quintela --- migration/multifd.h | 8 ++-- migration/multifd.c | 34 -- 2 files changed, 30 insertions(+), 12 deletions(-) diff --git a/migration/multifd.h

[PATCH v4 15/23] multifd: Use normal pages array on the recv side

Signed-off-by: Juan Quintela --- Rename num_normal_pages to total_normal_pages (peter) --- migration/multifd.h | 8 +-- migration/multifd-zlib.c | 8 +++ migration/multifd-zstd.c | 6 +++--- migration/multifd.c | 45 ++-- 4 files

Re: [PATCH v3 00/18] ppc/pnv: Extend the powernv10 machine

Hey, I suppose you`ll want to rebase these patches with the changes done in pnv-phb4 models and the exceptions cleanup. In that case I'll wait for a re-send before looking into these. I also read from the [1] [2] and [3] about PHB5. I guess we can introduce pnv-phb5 with the same premises

[PATCH v4 09/23] multifd: Make zlib use iov's

Signed-off-by: Juan Quintela Reviewed-by: Dr. David Alan Gilbert --- migration/multifd-zlib.c | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/migration/multifd-zlib.c b/migration/multifd-zlib.c index a2fec4d01d..71480c82bb 100644 --- a/migration/multifd-zlib.c +++

[PATCH v4 22/23] migration: Use multifd before we check for the zero page

So we use multifd to transmit zero pages. Signed-off-by: Juan Quintela --- migration/ram.c | 32 +++- 1 file changed, 31 insertions(+), 1 deletion(-) diff --git a/migration/ram.c b/migration/ram.c index 3536778e19..bdc7cec4cd 100644 --- a/migration/ram.c +++

[PATCH v4 07/23] multifd: Use proper maximum compression values

It happens that there are functions to calculate the worst possible compression size for a packet. Use them. Suggested-by: Dr. David Alan Gilbert Signed-off-by: Juan Quintela --- migration/multifd-zlib.c | 4 ++-- migration/multifd-zstd.c | 4 ++-- 2 files changed, 4 insertions(+), 4

Re: cirrus-ci: FreeBSD failure (lttng-ust package not found)

On Tue, Jan 11, 2022 at 02:11:14PM +0100, Philippe Mathieu-Daudé wrote: > Hi Alex, > > I am getting this failure for the x64-freebsd-*-build jobs [1, 2]: > > pkg: No packages available to install matching 'lttng-ust' have been > found in the repositories It was being installed fine on FreeBSD

[PATCH v4 05/23] migration: simplify do_compress_ram_page

The goto is not needed at all. Signed-off-by: Juan Quintela --- migration/ram.c | 11 +++ 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/migration/ram.c b/migration/ram.c index fa49d22e69..422c6bce28 100644 --- a/migration/ram.c +++ b/migration/ram.c @@ -1341,12 +1341,11

[PATCH v4 06/23] migration: Move ram_release_pages() call to save_zero_page_to_file()

We always need to call it when we find a zero page, so put it in a single place. Signed-off-by: Juan Quintela Reviewed-by: Peter Xu --- migration/ram.c | 21 ++--- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/migration/ram.c b/migration/ram.c index

[PATCH v4 19/23] multifd: Add property to enable/disable zero_page

Signed-off-by: Juan Quintela --- migration/migration.h | 3 +++ hw/core/machine.c | 4 +++- migration/migration.c | 11 +++ 3 files changed, 17 insertions(+), 1 deletion(-) diff --git a/migration/migration.h b/migration/migration.h index 8130b703eb..638cd89b6c 100644 ---

[PATCH v4 02/23] migration: We only need last_stage in two places

We only need last_stage in two places and we are passing it all around. Just add a field to RAMState that passes it. Signed-off-by: Juan Quintela Reviewed-by: Peter Xu --- Repeat subject (philmd suggestion) --- migration/ram.c | 41 ++--- 1 file changed,

[PATCH v4 04/23] migration: Remove masking for compression

Remove the mask in the call to ram_release_pages(). Nothing else does it, and if the offset has that bits set, we have a lot of trouble. Signed-off-by: Juan Quintela --- migration/ram.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/migration/ram.c b/migration/ram.c

[PATCH v4 14/23] multifd: Use normal pages array on the send side

We are only sending normal pages through multifd channels. Later on this series, we are going to also send zero pages. We are going to dectect if a page is zero or non zero in the multifd channel thread, not on the main thread. So we receive an array of pages page->offset[N] And we will end

[PATCH v4 01/23] migration: All this fields are unsigned

So printing it as %d is wrong. Notice that for the channel id, that is an uint8_t, but I changed it anyways for consistency. Signed-off-by: Juan Quintela Reviewed-by: Philippe Mathieu-Daudé Reviewed-by: Peter Xu --- migration/multifd-zlib.c | 20 ++-- migration/multifd-zstd.c

[PATCH v4 03/23] migration: ram_release_pages() always receive 1 page as argument

Remove the pages argument. And s/pages/page/ Signed-off-by: Juan Quintela Reviewed-by: Philippe Mathieu-Daudé Reviewed-by: Peter Xu -- - Use 1LL instead of casts (philmd) - Change the whole 1ULL for TARGET_PAGE_SIZE Signed-off-by: Juan Quintela --- migration/ram.c | 8 1 file

Re: [PATCH] hw/smbios: Add table 4 parameter, "processor-id"

On Thu, 6 Jan 2022 14:33:16 -0800 Patrick Venture wrote: > This parameter is to be used in the processor_id lower 32-bit entry in I'd call it processor_id_lo throughout the patch or if you prefer to keep current name, then make it support full QWORD as spec says. > the type 4 table. The

[PATCH v4 13/23] multifd: Unfold "used" variable by its value

Signed-off-by: Juan Quintela Reviewed-by: Dr. David Alan Gilbert --- migration/multifd.c | 8 +++- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/migration/multifd.c b/migration/multifd.c index 1d4885e1a0..e5b1fa5015 100644 --- a/migration/multifd.c +++ b/migration/multifd.c

Re: [PATCH 4/5] usb: allow max 8192 bytes for desc

On 1/5/22 08:25, zhenwei pi wrote: > > On 1/4/22 11:22 PM, Philippe Mathieu-Daudé wrote: >> On 27/12/21 15:27, zhenwei pi wrote: >>> A device of USB video class usually uses larger desc structure, so >>> use larger buffer to avoid failure. >>> >>> Signed-off-by: zhenwei pi >>> --- >>>  

[PATCH v5 4/5] ppc/pnv: Introduce user creatable pnv-phb4 devices

This patch introduces pnv-phb4 user creatable devices that are created in a similar manner as pnv-phb3 devices, allowing the user to interact with the PHBs directly instead of creating PCI Express Controllers that will create a certain amount of PHBs per controller index. We accomplish this by

[PATCH v4 00/23] Migration: Transmit and detect zero pages in the multifd threads

Hi In this version - Rebase to latest - Address all comments from previous versions - code cleanup Please review. [v2] This is a rebase against last master. And the reason for resend is to configure properly git-publish and hope this time that git-publish send all the patches. Please, review.

[PATCH v4 12/23] multifd: Use a single writev on the send side

Until now, we wrote the packet header with write(), and the rest of the pages with writev(). Just increase the size of the iovec and do a single writev(). Signed-off-by: Juan Quintela Reviewed-by: Dr. David Alan Gilbert --- migration/multifd.c | 20 1 file changed, 8

Re: [PATCH v1] util/oslib-posix: Fix missing unlock in the error path of os_mem_prealloc()

> We're missing an unlock in case installing the signal handler failed. > Fortunately, we barely see this error in real life. > > Fixes: a960d6642d39 ("util/oslib-posix: Support concurrent os_mem_prealloc() > invocation") > Fixes: CID 1468941 > Cc: Paolo Bonzini > Cc: Michael S. Tsirkin > Cc:

Re: [PATCH] clock-vmstate: Add missing END_OF_LIST

On 1/11/22 11:19, Dr. David Alan Gilbert (git) wrote: > From: "Dr. David Alan Gilbert" > > Add the missing VMSTATE_END_OF_LIST to vmstate_muldiv > > Fixes: 99abcbc7600 ("clock: Provide builtin multiplier/divider") Doh I missed this too :( Reviewed-by: Philippe Mathieu-Daudé > Signed-off-by:

Re: [PATCH 8/8] target/ppc: 405: System call exception cleanup

David Gibson writes: > On Mon, Jan 10, 2022 at 03:15:46PM -0300, Fabiano Rosas wrote: >> There's no sc 1. >> >> We also only used env->nip because of the vhyp code, so change to >> 'vector' now. > > I don't think this is right. The point with the env->nip change is > changing the PC as it

[PATCH v4 10/23] multifd: Make zstd use iov's

Signed-off-by: Juan Quintela Reviewed-by: Dr. David Alan Gilbert --- migration/multifd-zstd.c | 8 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/migration/multifd-zstd.c b/migration/multifd-zstd.c index 97c08367d0..bd393aee0d 100644 --- a/migration/multifd-zstd.c +++

Re: [PATCH v1] libvhost-user: Map shared RAM with MAP_NORESERVE to support virtio-mem with hugetlb

On 11.01.22 13:39, David Hildenbrand wrote: > For fd-based shared memory, MAP_NORESERVE is only effective for hugetlb, > otherwise it's ignored. Older Linux versions that didn't support > reservation of huge pages ignored MAP_NORESERVE completely. > > The first client to mmap a hugetlb fd without

Re: [PATCH] usb: allow max 8192 bytes for desc

On 1/11/22 13:27, zhenwei pi wrote: > On 1/11/22 8:25 PM, Daniel P. Berrangé wrote: >> On Tue, Jan 11, 2022 at 12:21:42PM +, Peter Maydell wrote: >>> On Tue, 11 Jan 2022 at 10:54, zhenwei pi >>> wrote: A device of USB video class usually uses larger desc structure, so use

[PATCH v3 6/6] migration: Move ram_release_pages() call to save_zero_page_to_file()

We always need to call it when we find a zero page, so put it in a single place. Signed-off-by: Juan Quintela Reviewed-by: Peter Xu --- migration/ram.c | 21 ++--- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/migration/ram.c b/migration/ram.c index

[PATCH v4 16/23] multifd: recv side only needs the RAMBlock host address

So we can remove the MultiFDPages. Signed-off-by: Juan Quintela Reviewed-by: Dr. David Alan Gilbert --- migration/multifd.h | 4 ++-- migration/multifd-zlib.c | 2 +- migration/multifd-zstd.c | 2 +- migration/multifd.c | 7 ++- 4 files changed, 6 insertions(+), 9 deletions(-)

Re: [RFC PATCH v2 2/6] audio/coreaudio: Remove a deprecation warning on macOS 12

On Dienstag, 11. Januar 2022 13:35:05 CET Christian Schoenebeck wrote: > Curiousity was stronger: the original clang behaviour was as I explained: > https://github.com/llvm-mirror/clang/commit/0a0d2b179085a52c10402feebeb6db8b >

[PATCH v4 11/23] multifd: Remove send_write() method

Everything use now iov's. Signed-off-by: Juan Quintela --- migration/multifd.h | 2 -- migration/multifd-zlib.c | 17 - migration/multifd-zstd.c | 17 - migration/multifd.c | 20 ++-- 4 files changed, 2 insertions(+), 54 deletions(-)

[PATCH v3 0/6] migration: misc cleanups

Hi Changes since v2: - rebase - address comments - use TARGET_PAGE_SIZE and remove casts - remove TARGET_PAGE_MASK from compression code. Nothing else use them. And if offsets are not aligned, we are in big trouble. Please, review. Changes since v1: - Add reviewed tags for reviewed patches

[PATCH v3 3/6] migration: ram_release_pages() always receive 1 page as argument

Remove the pages argument. And s/pages/page/ Signed-off-by: Juan Quintela Reviewed-by: Philippe Mathieu-Daudé Reviewed-by: Peter Xu -- - Use 1LL instead of casts (philmd) - Change the whole 1ULL for TARGET_PAGE_SIZE Signed-off-by: Juan Quintela --- migration/ram.c | 8 1 file

[PATCH v3 4/6] migration: Remove masking for compression

Remove the mask in the call to ram_release_pages(). Nothing else does it, and if the offset has that bits set, we have a lot of trouble. Signed-off-by: Juan Quintela --- migration/ram.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/migration/ram.c b/migration/ram.c

Re: [PATCH v2 19/37] console: save current scanout details

On 2022/01/11 17:23, Marc-André Lureau wrote: Hi Akihiko On Tue, Jan 11, 2022 at 7:30 AM Akihiko Odaki wrote: Hi, I found this brings an inconsistency and a flaw to scanout semantics and think the inconsistency should be fixed or this should be reverted before the next release comes up.

[PATCH v3 2/6] migration: We only need last_stage in two places

We only need last_stage in two places and we are passing it all around. Just add a field to RAMState that passes it. Signed-off-by: Juan Quintela Reviewed-by: Peter Xu --- Repeat subject (philmd suggestion) --- migration/ram.c | 41 ++--- 1 file changed,

[PATCH v3 1/6] migration: All this fields are unsigned

So printing it as %d is wrong. Notice that for the channel id, that is an uint8_t, but I changed it anyways for consistency. Signed-off-by: Juan Quintela Reviewed-by: Philippe Mathieu-Daudé Reviewed-by: Peter Xu --- migration/multifd-zlib.c | 20 ++-- migration/multifd-zstd.c

[PATCH v1] libvhost-user: Map shared RAM with MAP_NORESERVE to support virtio-mem with hugetlb

For fd-based shared memory, MAP_NORESERVE is only effective for hugetlb, otherwise it's ignored. Older Linux versions that didn't support reservation of huge pages ignored MAP_NORESERVE completely. The first client to mmap a hugetlb fd without MAP_NORESERVE will trigger reservation of huge pages

[PATCH v3 5/6] migration: simplify do_compress_ram_page

The goto is not needed at all. Signed-off-by: Juan Quintela --- migration/ram.c | 11 +++ 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/migration/ram.c b/migration/ram.c index fa49d22e69..422c6bce28 100644 --- a/migration/ram.c +++ b/migration/ram.c @@ -1341,12 +1341,11

Re: Re: Re: [PATCH] usb: allow max 8192 bytes for desc

On 1/11/22 8:38 PM, Daniel P. Berrangé wrote: On Tue, Jan 11, 2022 at 08:27:35PM +0800, zhenwei pi wrote: On 1/11/22 8:25 PM, Daniel P. Berrangé wrote: On Tue, Jan 11, 2022 at 12:21:42PM +, Peter Maydell wrote: On Tue, 11 Jan 2022 at 10:54, zhenwei pi wrote: A device of USB video

Re: Re: [PATCH] usb: allow max 8192 bytes for desc

On Tue, Jan 11, 2022 at 08:27:35PM +0800, zhenwei pi wrote: > > > On 1/11/22 8:25 PM, Daniel P. Berrangé wrote: > > On Tue, Jan 11, 2022 at 12:21:42PM +, Peter Maydell wrote: > > > On Tue, 11 Jan 2022 at 10:54, zhenwei pi wrote: > > > > > > > > A device of USB video class usually uses

Re: [PATCH 4/5] usb: allow max 8192 bytes for desc

On Mon, Dec 27, 2021 at 10:27:33PM +0800, zhenwei pi wrote: > A device of USB video class usually uses larger desc structure, so > use larger buffer to avoid failure. > > Signed-off-by: zhenwei pi > --- > hw/usb/desc.c | 15 --- > hw/usb/desc.h | 1 + > 2 files changed, 9

Re: [RFC PATCH v2 2/6] audio/coreaudio: Remove a deprecation warning on macOS 12

On Montag, 10. Januar 2022 22:05:43 CET Christian Schoenebeck wrote: > On Montag, 10. Januar 2022 21:39:28 CET Akihiko Odaki wrote: > > On 2022/01/11 5:22, Christian Schoenebeck wrote: > > > On Montag, 10. Januar 2022 20:01:40 CET Akihiko Odaki wrote: > > >> On 2022/01/11 3:46, Christian

Re: Re: [PATCH] usb: allow max 8192 bytes for desc

On 1/11/22 8:25 PM, Daniel P. Berrangé wrote: On Tue, Jan 11, 2022 at 12:21:42PM +, Peter Maydell wrote: On Tue, 11 Jan 2022 at 10:54, zhenwei pi wrote: A device of USB video class usually uses larger desc structure, so use larger buffer to avoid failure. (dev-video.c is ready)

Re: [PATCH] usb: allow max 8192 bytes for desc

On Tue, Jan 11, 2022 at 12:21:42PM +, Peter Maydell wrote: > On Tue, 11 Jan 2022 at 10:54, zhenwei pi wrote: > > > > A device of USB video class usually uses larger desc structure, so > > use larger buffer to avoid failure. (dev-video.c is ready) > > > > Allocating memory dynamically by

Re: [PATCH] usb: allow max 8192 bytes for desc

On Tue, 11 Jan 2022 at 10:54, zhenwei pi wrote: > > A device of USB video class usually uses larger desc structure, so > use larger buffer to avoid failure. (dev-video.c is ready) > > Allocating memory dynamically by g_malloc of the orignal version of > this change, Philippe suggested just using

Re: [PATCH] usb: allow max 8192 bytes for desc

On 1/11/22 11:49, zhenwei pi wrote: > A device of USB video class usually uses larger desc structure, so > use larger buffer to avoid failure. (dev-video.c is ready) > > Allocating memory dynamically by g_malloc of the orignal version of > this change, Philippe suggested just using the stack.

Re: [PATCH] clock-vmstate: Add missing END_OF_LIST

On Tue, 11 Jan 2022 at 10:19, Dr. David Alan Gilbert (git) wrote: > > From: "Dr. David Alan Gilbert" > > Add the missing VMSTATE_END_OF_LIST to vmstate_muldiv > > Fixes: 99abcbc7600 ("clock: Provide builtin multiplier/divider") > Signed-off-by: Dr. David Alan Gilbert > --- >

Re: [PATCH v4 7/7] tcg/arm: Support raising sigbus for user-only

On Sat, 8 Jan 2022 at 06:33, Richard Henderson wrote: > > Signed-off-by: Richard Henderson > --- > +static void tcg_out_test_alignment(TCGContext *s, bool is_ld, TCGReg addrlo, > + TCGReg addrhi, unsigned a_bits) > @@ -1709,6 +1778,10 @@ static void

[PATCH] Fix null pointer dereference in util/fdmon-epoll.c

Orginal qemu commit hash: de3f5223fa4cf8bfc5e3fe1fd495ddf468edcdf7 In util/fdmon-epoll.c, function fdmon_epoll_update, variable "old_node" maybe NULL with the condition, while it is directly used in the statement and may lead to null pointer dereferencen problem. Variable "r" in the condition is

[PATCH v1] util/oslib-posix: Fix missing unlock in the error path of os_mem_prealloc()

We're missing an unlock in case installing the signal handler failed. Fortunately, we barely see this error in real life. Fixes: a960d6642d39 ("util/oslib-posix: Support concurrent os_mem_prealloc() invocation") Fixes: CID 1468941 Cc: Paolo Bonzini Cc: Michael S. Tsirkin Cc: Pankaj Gupta Cc:

Re: [PATCH v4 6/7] tcg/arm: Reserve a register for guest_base

On Sat, 8 Jan 2022 at 06:33, Richard Henderson wrote: > > Reserve a register for the guest_base using aarch64 for reference. > By doing so, we do not have to recompute it for every memory load. > > Signed-off-by: Richard Henderson > --- > tcg/arm/tcg-target.c.inc | 39

Re: [PATCH v4 5/7] tcg/arm: Support unaligned access for softmmu

On Sat, 8 Jan 2022 at 06:33, Richard Henderson wrote: > > From armv6, the architecture supports unaligned accesses. > All we need to do is perform the correct alignment check > in tcg_out_tlb_read. > > Signed-off-by: Richard Henderson > --- > tcg/arm/tcg-target.c.inc | 39

Re: [PULL 0/2] SD/MMC patches for 2022-01-08

On Mon, 10 Jan 2022 at 19:20, Philippe Mathieu-Daudé wrote: > > On 1/10/22 17:02, Peter Maydell wrote: > > On Sat, 8 Jan 2022 at 21:59, Philippe Mathieu-Daudé wrote: > >> > >> Hi Richard, > >> > >> This is the SD/MMC PR that ought to be sent previously. > >> > >> The following changes since

Re: [PULL v2 00/10] virtio: revert config interrupt changes

On Mon, 10 Jan 2022 at 21:04, Michael S. Tsirkin wrote: > > Changes from v1: > added tags to commit logs. > > The following changes since commit ca745d2277496464b54fd832c15c45d0227325bb: > > tests: acpi: Add updated TPM related tables (2022-01-07 19:30:13 -0500) > > are available in the Git

Re: [PATCH v4 4/7] tcg/arm: Check alignment for ldrd and strd

On Sat, 8 Jan 2022 at 06:33, Richard Henderson wrote: > > We will shortly allow the use of unaligned memory accesses, > and these require proper alignment. Use get_alignment_bits > to verify and remove USING_SOFTMMU. > > Signed-off-by: Richard Henderson > --- > tcg/arm/tcg-target.c.inc | 22

Re: [PATCH v4 3/7] tcg/arm: Remove use_armv6_instructions

On Sat, 8 Jan 2022 at 06:33, Richard Henderson wrote: > > This is now always true, since we require armv6. > > Signed-off-by: Richard Henderson Reviewed-by: Peter Maydell thanks -- PMM

Re: [PATCH v4 2/7] tcg/arm: Remove use_armv5t_instructions

On Sat, 8 Jan 2022 at 06:33, Richard Henderson wrote: > > This is now always true, since we require armv6. > > Signed-off-by: Richard Henderson Reviewed-by: Peter Maydell thanks -- PMM

Re: [PATCH v4 1/7] tcg/arm: Drop support for armv4 and armv5 hosts

On Sat, 8 Jan 2022 at 06:33, Richard Henderson wrote: > > Support for unaligned accesses is difficult for pre-v6 hosts. > While debian still builds for armv4, we cannot use a compile > time test, so test the architecture at runtime and error out. > > Signed-off-by: Richard Henderson

Re: [PATCH 2/2] block/rbd: workaround for ceph issue #53784

Hi Peter, On Tue, Jan 11, 2022 at 10:10:16AM +0100, Peter Lieven wrote: Hi Stefano, thanks for the feedback. Please note that you also need the other patch or you will sooner or later run into another assertion as soon as rbd snapshots are involved. Yep, I tested with the entire series

Re: [PATCH v2] block-backend: prevent dangling BDS pointers across aio_poll()

On Wed, Dec 15, 2021 at 04:31:26PM +0100, Kevin Wolf wrote: > Am 15.12.2021 um 12:28 hat Stefan Hajnoczi geschrieben: > > On Tue, Dec 14, 2021 at 03:59:49PM +0100, Kevin Wolf wrote: > > > Am 14.12.2021 um 15:35 hat Stefan Hajnoczi geschrieben: > > > > The BlockBackend root child can change when

Re: [PATCH v1 2/2] hw/i386: Make pic a property of common x86 base machine type

On Tue, Jan 11, 2022 at 03:35:28PM +0800, Xiaoyao Li wrote: > Legacy PIC (8259) cannot be supported for TDX guests since TDX module > doesn't allow directly interrupt injection. Using posted interrupts > for the PIC is not a viable option as the guest BIOS/kernel will not > do EOI for PIC IRQs,

Re: [PATCH v1 1/2] hw/i386: Make pit a property of common x86 base machine type

On Tue, Jan 11, 2022 at 03:35:27PM +0800, Xiaoyao Li wrote: > Both pc and microvm have pit property individually. Let's just make it > the property of common x86 base machine type. > > Signed-off-by: Xiaoyao Li > --- > hw/i386/microvm.c | 27 +-- > hw/i386/pc.c

[PATCH] usb: allow max 8192 bytes for desc

A device of USB video class usually uses larger desc structure, so use larger buffer to avoid failure. (dev-video.c is ready) Allocating memory dynamically by g_malloc of the orignal version of this change, Philippe suggested just using the stack. Test the two versions of qemu binary, the size of

Re: [PATCH v2 1/9] hw/nvram: Restrict stub to sysemu and tools

On 1/7/22 01:17, Richard Henderson wrote: > On 1/4/22 12:54 AM, Philippe Mathieu-Daudé wrote: >> +if have_system or have_tools > ... >>   if have_system >> +  stub_ss.add(files('fw_cfg.c')) > > Disconnect in tests? (I assume s/tests/stubs/). fw_cfg-interface.c exposes the fw_cfg QOM interface,

Re: [PATCH 0/3] virtio-iommu: Support VIRTIO_IOMMU_F_BYPASS_CONFIG

Hi Jean, On 1/11/22 11:13 AM, Jean-Philippe Brucker wrote: > Hi Eric, > > On Tue, Jan 11, 2022 at 10:02:12AM +0100, Eric Auger wrote: >> Hi Jean, Michael, >> >> On 9/30/21 8:50 PM, Jean-Philippe Brucker wrote: >>> Replace the VIRTIO_IOMMU_F_BYPASS feature with >>> VIRTIO_IOMMU_F_BYPASS_CONFIG,

Re: [PATCH] clock-vmstate: Add missing END_OF_LIST

On Tue, 11 Jan 2022 at 10:19, Dr. David Alan Gilbert (git) wrote: > > From: "Dr. David Alan Gilbert" > > Add the missing VMSTATE_END_OF_LIST to vmstate_muldiv > > Fixes: 99abcbc7600 ("clock: Provide builtin multiplier/divider") > Signed-off-by: Dr. David Alan Gilbert > --- >

Re: "Startup" meeting (was Re: Meeting today?)

On Tue, Jan 11, 2022 at 11:20:54AM +0100, Philippe Mathieu-Daudé wrote: > Hi, > > Just checking in, this call is scheduled for today, 3pm CEST, right? > > Here is the KVM call calendar: >

Re: [PATCH v4 4/5] ppc/pnv: Introduce user creatable pnv-phb4 devices

On 1/11/22 06:56, Cédric Le Goater wrote: On 1/11/22 01:55, Daniel Henrique Barboza wrote: This patch introduces pnv-phb4 user creatable devices that are created in a similar manner as pnv-phb3 devices, allowing the user to interact with the PHBs directly instead of creating PCI Express

Re: "Startup" meeting (was Re: Meeting today?)

That is my understanding… See you there! Cheers Mark. > On 11 Jan 2022, at 11:20, Philippe Mathieu-Daudé wrote: > > Hi, > > Just checking in, this call is scheduled for today, 3pm CEST, right? > > Here is the KVM call calendar: >

Re: "Startup" meeting (was Re: Meeting today?)

Hi, Just checking in, this call is scheduled for today, 3pm CEST, right? Here is the KVM call calendar: https://www.google.com/calendar/embed?src=dG9iMXRqcXAzN3Y4ZXZwNzRoMHE4a3BqcXNAZ3JvdXAuY2FsZW5kYXIuZ29vZ2xlLmNvbQ On 1/6/22 12:23, Daniel P. Berrangé wrote: > No one objected, so I think we

[PATCH] clock-vmstate: Add missing END_OF_LIST

From: "Dr. David Alan Gilbert" Add the missing VMSTATE_END_OF_LIST to vmstate_muldiv Fixes: 99abcbc7600 ("clock: Provide builtin multiplier/divider") Signed-off-by: Dr. David Alan Gilbert --- hw/core/clock-vmstate.c | 1 + 1 file changed, 1 insertion(+) diff --git a/hw/core/clock-vmstate.c

Re: [PATCH 0/3] virtio-iommu: Support VIRTIO_IOMMU_F_BYPASS_CONFIG

Hi Eric, On Tue, Jan 11, 2022 at 10:02:12AM +0100, Eric Auger wrote: > Hi Jean, Michael, > > On 9/30/21 8:50 PM, Jean-Philippe Brucker wrote: > > Replace the VIRTIO_IOMMU_F_BYPASS feature with > > VIRTIO_IOMMU_F_BYPASS_CONFIG, which enables a config space bit to switch > > global bypass on and

Re: [PULL v4 00/18] Build system and KVM changes for 2021-12-23

On Mon, 10 Jan 2022 at 15:01, Paolo Bonzini wrote: > > The following changes since commit afe33262585565b64df706c62b4b0f6e0ad30c71: > > Merge tag 'pull-riscv-to-apply-20220108' of github.com:alistair23/qemu into > staging (2022-01-07 22:09:24 -0800) > > are available in the Git repository at:

Re: [PATCH v4 5/5] ppc/pnv: turn pnv_phb4_update_regions() into static

On 1/11/22 01:55, Daniel Henrique Barboza wrote: Its only callers are inside pnv_phb4.c. Signed-off-by: Daniel Henrique Barboza Reviewed-by: Cédric Le Goater Thanks, C. --- hw/pci-host/pnv_phb4.c | 52 +- include/hw/pci-host/pnv_phb4.h | 1 -

Re: [PATCH v4 4/5] ppc/pnv: Introduce user creatable pnv-phb4 devices

On 1/11/22 01:55, Daniel Henrique Barboza wrote: This patch introduces pnv-phb4 user creatable devices that are created in a similar manner as pnv-phb3 devices, allowing the user to interact with the PHBs directly instead of creating PCI Express Controllers that will create a certain amount of

Re: [PATCH v4 2/5] ppc/pnv: move PHB4 XSCOM init to phb4_realize()

On 1/11/22 01:55, Daniel Henrique Barboza wrote: The 'stack->phb_regs_mr' PHB4 passthrough XSCOM initialization relies on 'stack->phb' being not NULL. Moving 'stack->phb_regs_mr' region_init() and add_subregion() to phb4_realize() time is a natural thing to do since it's strictly PHB related.

Re: [PATCH v4 3/5] ppc/pnv: turn 'phb' into a pointer in struct PnvPhb4PecStack

On 1/11/22 01:55, Daniel Henrique Barboza wrote: At this moment, stack->phb is the plain PnvPHB4 device itself instead of a pointer to the device. This will present a problem when adding user creatable devices because we can't deal with this struct and the realize() callback from the user

Re: [PATCH v4 07/14] vfio-user: run vfio-user context

On Mon, Jan 10, 2022 at 05:56:25PM +, John Levon wrote: > On Thu, Jan 06, 2022 at 01:35:32PM +, Stefan Hajnoczi wrote: > > > > > >> +static void vfu_object_attach_ctx(void *opaque) > > > > >> +{ > > > > >> +VfuObject *o = opaque; > > > > >> +GPollFD pfds[1]; > > > > >> +int

Re: [PATCH v4 1/5] ppc/pnv: set phb4 properties in stk_realize()

On 1/11/22 01:55, Daniel Henrique Barboza wrote: Moving all phb4 properties setup to stk_realize() keeps this logic in a single place instead of having it scattered between stk_realize() and pec_realize(). 'phb->index' can be retrieved using stack->stack_no and pnv_phb4_pec_get_phb_id(),

RE: [PATCH 1/1] util: adjust coroutine pool size to virtio block queue

> On Thu, Jan 06, 2022 at 05:20:57PM +0900, Hiroki Narukawa wrote: > > Phil, thanks for notifying me. > > > Coroutine pool size was 64 from long ago, and the basis was organized in > > the commit message in c740ad92. > > > > At that time, virtio-blk queue-size and num-queue were not

Re: [PATCH v3 0/2] Aspeed I3C device model

On 1/11/22 09:45, Troy Lee wrote: This series of patch introduce a dummy implemenation of aspeed i3c model, and it provide just enough information for guest machine. However, the driver probing is still failed, but it will not cause kernel panic. v3: - Remove unused AspeedI3CClass - Refine

[PATCH 0/1 v2] Patch to adjust coroutine pool size adaptively

Resending patch with fixing atomic access method to use qatomic_read(). We encountered random disk IO performance drop since qemu-5.0.0, and this patch fixes it. Commit message in c740ad92 implied to adjust coroutine pool size adaptively, so I tried to implement this. Could you review this

[PATCH 1/1] util: adjust coroutine pool size to virtio block queue

Coroutine pool size was 64 from long ago, and the basis was organized in the commit message in c740ad92. At that time, virtio-blk queue-size and num-queue were not configuable, and equivalent values were 128 and 1. Coroutine pool size 64 was fine then. Later queue-size and num-queue got

Re: [PATCH v3 2/2] This patch includes i3c instance in ast2600 soc.

On 1/11/22 09:45, Troy Lee wrote: v3: - Remove unrelated changes to SPI2 address - Remove controller irq line v2: Rebase to mainline QEMU Signed-off-by: Troy Lee Reviewed-by: Cédric Le Goater You should consider updating the documentation :

Re: [PATCH v2 1/5] libvhost-user: Add vu_rem_mem_reg input validation

On Mon, Jan 10, 2022 at 07:43:06PM +, Raphael Norwitz wrote: > On Mon, Jan 10, 2022 at 04:36:34AM -0500, Michael S. Tsirkin wrote: > > On Thu, Jan 06, 2022 at 06:47:26AM +, Raphael Norwitz wrote: > > > Signed-off-by: Raphael Norwitz > > > > > > Raphael any chance you can add a bit more

Re: [PATCH] softmmu/device_tree: Remove redundant pointer assignment

On Tue, Jan 11, 2022 at 11:27:58AM +0800, Yanan Wang wrote: > The pointer assignment "const char *p = path;" in function > qemu_fdt_add_path is unnecessary. Let's remove it and just > use the "path" passed in. No functional change. > > Suggested-by: Richard Henderson > Signed-off-by: Yanan Wang

Re: [PATCH 2/2] block/rbd: workaround for ceph issue #53784

Am 10.01.22 um 15:18 schrieb Stefano Garzarella: > On Mon, Jan 10, 2022 at 12:41:54PM +0100, Peter Lieven wrote: >> librbd had a bug until early 2022 that affected all versions of ceph that >> supported fast-diff. This bug results in reporting of incorrect offsets >> if the offset parameter to

Re: [PATCH v3 1/2] Introduce a dummy AST2600 I3C model.

On 1/11/22 09:45, Troy Lee wrote: Aspeed 2600 SDK enables I3C support by default. The I3C driver will try to reset the device controller and setup through device address table register. This dummy model response these register with default value listed on ast2600v10 datasheet chapter 54.2. If

Re: [PATCH] target/ppc: Fix e6500 boot

>From: "Cédric Le Goater" c...@kaod.org >To: "ma...@locati.it" ma...@locati.it, bala...@eik.bme.hu >Cc: faro...@linux.ibm.com, qemu-devel@nongnu.org, qemu-...@nongnu.org, >danielhb...@gmail.com >Date: Mon, 10 Jan 2022 09:04:39 +0100 >Subject: Re: [PATCH] target/ppc: Fix e6500 boot > >Hello Mario,

Re: [PATCH 0/3] virtio-iommu: Support VIRTIO_IOMMU_F_BYPASS_CONFIG

Hi Jean, Michael, On 9/30/21 8:50 PM, Jean-Philippe Brucker wrote: > Replace the VIRTIO_IOMMU_F_BYPASS feature with > VIRTIO_IOMMU_F_BYPASS_CONFIG, which enables a config space bit to switch > global bypass on and off. > > Add a boot-bypass option, which defaults to 'on' to be in line with >

Re: [PATCH v2] linux-user: Fix clang warning for nios2-linux-user code

Le 11/01/2022 à 09:29, Peter Maydell a écrit : The clang in Ubuntu 18.04 (10.0.0-4ubuntu1) produces a warning on the code added in commit f5ef0e518d03 where we use a shifted expression in a boolean context: ../../linux-user/elfload.c:2423:16: error: converting the result of '<<' to a boolean

[PATCH v3 2/2] This patch includes i3c instance in ast2600 soc.

v3: - Remove unrelated changes to SPI2 address - Remove controller irq line v2: Rebase to mainline QEMU Signed-off-by: Troy Lee --- hw/arm/aspeed_ast2600.c | 16 include/hw/arm/aspeed_soc.h | 3 +++ 2 files changed, 19 insertions(+) diff --git a/hw/arm/aspeed_ast2600.c

[PATCH v3 1/2] Introduce a dummy AST2600 I3C model.

Aspeed 2600 SDK enables I3C support by default. The I3C driver will try to reset the device controller and setup through device address table register. This dummy model response these register with default value listed on ast2600v10 datasheet chapter 54.2. If the device address table register

[PATCH v3 0/2] Aspeed I3C device model

This series of patch introduce a dummy implemenation of aspeed i3c model, and it provide just enough information for guest machine. However, the driver probing is still failed, but it will not cause kernel panic. v3: - Remove unused AspeedI3CClass - Refine memory region - Refine register reset -

<    1   2   3   4   >