Migration 20230209 patches
Subsystems (3)
The following changes since commit 417296c8d8588f782018d01a317f88957e9786d6:
tests/qtest/netdev-socket: Raise connection timeout to 60 seconds (2023-02-09 11:23:53 +0000)
are available in the Git repository at:
https://gitlab.com/juan.quintela/qemu.git tags/migration-20230209-pull-request
for you to fetch changes up to 858191aebda251a4d1e3bc77b238096673241cdd:
migration: Postpone postcopy preempt channel to be after main (2023-02-09 21:26:02 +0100)
----------------------------------------------------------------
Migration Pull request
Hi
This are all the reviewed patches for migration:
- AVX512 support for xbzrle (Ling Xu)
- /dev/userfaultd support (Peter Xu)
- Improve ordering of channels (Peter Xu)
- multifd cleanups (Li Zhang)
- Remove spurious files from last merge (me) Rebase makes that to you
- Fix mixup between state_pending_{exact,estimate} (me)
- Cache RAM size during migration (me)
- cleanup several functions (me)
Please apply.
----------------------------------------------------------------
Juan Quintela (7):
migration: Remove spurious files
migration: Simplify ram_find_and_save_block()
migration: Make find_dirty_block() return a single parameter
migration: Split ram_bytes_total_common() in two functions
migration: Calculate ram size once
migration: Make ram_save_target_page() a pointer
migration: I messed state_pending_exact/estimate
Li Zhang (2):
multifd: cleanup the function multifd_channel_connect
multifd: Remove some redundant code
Peter Xu (6):
linux-headers: Update to v6.1
util/userfaultfd: Support /dev/userfaultfd
migration: Rework multi-channel checks on URI
migration: Cleanup postcopy_preempt_setup()
migration: Add a semaphore to count PONGs
migration: Postpone postcopy preempt channel to be after main
ling xu (2):
AVX512 support for xbzrle_encode_buffer
Update bench-code for addressing CI problem
MAINTAINERS | 1 + .../x86_64-quintela-devices.mak | 7 - .../x86_64-quintela2-devices.mak | 6 - meson.build | 17 + include/standard-headers/drm/drm_fourcc.h | 34 +- include/standard-headers/linux/ethtool.h | 63 +- include/standard-headers/linux/fuse.h | 6 +- .../linux/input-event-codes.h | 1 + include/standard-headers/linux/virtio_blk.h | 19 + linux-headers/asm-generic/hugetlb_encode.h | 26 +- linux-headers/asm-generic/mman-common.h | 2 + linux-headers/asm-mips/mman.h | 2 + linux-headers/asm-riscv/kvm.h | 4 + linux-headers/linux/kvm.h | 1 + linux-headers/linux/psci.h | 14 + linux-headers/linux/userfaultfd.h | 4 + linux-headers/linux/vfio.h | 142 ++ migration/migration.h | 15 +- migration/postcopy-ram.h | 4 +- migration/xbzrle.h | 4 + migration/migration.c | 122 +- migration/multifd.c | 70 +- migration/postcopy-ram.c | 31 +- migration/ram.c | 136 +- migration/savevm.c | 56 +- migration/xbzrle.c | 124 ++ tests/bench/xbzrle-bench.c | 473 ++++++ tests/unit/test-xbzrle.c | 42 +- util/userfaultfd.c | 32 + meson_options.txt | 2 + migration/multifd.c.orig | 1274 ----------------- scripts/meson-buildoptions.sh | 3 + tests/bench/meson.build | 4 + util/trace-events | 1 + 34 files changed, 1257 insertions(+), 1485 deletions(-) delete mode 100644 configs/devices/x86_64-softmmu/x86_64-quintela-devices.mak delete mode 100644 configs/devices/x86_64-softmmu/x86_64-quintela2-devices.mak create mode 100644 tests/bench/xbzrle-bench.c delete mode 100644 migration/multifd.c.orig
Signature · 1 line
2.39.1
[PULL 02/17] multifd: cleanup the function multifd_channel_connect
From: Li Zhang <[email protected]>
Cleanup multifd_channel_connect
--- migration/multifd.c | 43 +++++++++++++++++++++---------------------- 1 file changed, 21 insertions(+), 22 deletions(-)
migration/multifd.c +21 -22
@@ -843,30 +843,29 @@ static bool multifd_channel_connect(MultiFDSendParams *p, ioc,object_get_typename (OBJECT (ioc)), migrate_get_current()->hostname, error); -if (!error) { -if (migrate_channel_requires_tls_upgrade (ioc)) { -multifd_tls_channel_connect (p, ioc, &error); -if (!error) { -/* -* tls_channel_connect will call back to this -* function after the TLS handshake, -* so we mustn't call multifd_send_thread until then -*/ -return true; - }else { -return false; - } +if (error) { +return false; + } +if (migrate_channel_requires_tls_upgrade (ioc)) { +multifd_tls_channel_connect (p, ioc, &error); +if (!error) { +/* +* tls_channel_connect will call back to this +* function after the TLS handshake, +* so we mustn't call multifd_send_thread until then +*/ +return true; }else { -migration_ioc_register_yank (ioc); - p->registered_yank = true; - p->c = ioc; -qemu_thread_create (&p->thread, p->name, multifd_send_thread, p, -QEMU_THREAD_JOINABLE ); - } -return true; +return false; + } + }else { +migration_ioc_register_yank (ioc); + p->registered_yank = true; + p->c = ioc; +qemu_thread_create (&p->thread, p->name, multifd_send_thread, p, +QEMU_THREAD_JOINABLE ); } - -return false; +return true; } static void multifd_new_send_channel_cleanup(MultiFDSendParams *p,
Signature · 1 line
2.39.1
[PULL 03/17] multifd: Remove some redundant code
In reply to Juan QuintelaFrom: Li Zhang <[email protected]>
Clean up some unnecessary code
--- migration/multifd.c | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-)
migration/multifd.c +4 -11
@@ -892,19 +892,15 @@ static void multifd_new_send_channel_async(QIOTask *task, gpointer opaque)Error *local_err =NULL ;trace_multifd_new_send_channel_async (p->id); -if (qio_task_propagate_error (task, &local_err)) { - goto cleanup; - }else { +if (!qio_task_propagate_error (task, &local_err)) { p->c =QIO_CHANNEL (sioc);qio_channel_set_delay (p->c, false); p->running = true; -if (!multifd_channel_connect (p, sioc, local_err)) { - goto cleanup; +if (multifd_channel_connect (p, sioc, local_err)) { +return ; } -return ; } -cleanup:multifd_new_send_channel_cleanup (p, sioc, local_err); }
@@ -1115,10 +1111,7 @@ static void *multifd_recv_thread(void *opaque) ret =qio_channel_read_all_eof (p->c, (void *)p->packet, p->packet_len, &local_err); -if (ret ==0 ) {/* EOF */ -break ; - } -if (ret ==-1 ) {/* Error */ +if (ret ==0 || ret ==-1 ) {/* 0: EOF -1: Error */ break ; }
Signature · 1 line
2.39.1
[PULL 08/17] migration: Split ram_bytes_total_common() in two functions
In reply to Juan QuintelaIt is just a big if in the middle of the function, and we need two functions anways.
---
Reindent to make Phillipe happy (and CODING_STYLE)
--- migration/ram.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-)
migration/ram.c +14 -11
@@ -2601,28 +2601,30 @@ void acct_update_position(QEMUFile *f, size_t size, bool zero) } } -static uint64_t ram_bytes_total_common (bool count_ignored) +static uint64_t ram_bytes_total_with_ignored (void ) {RAMBlock *block;uint64_t total =0 ;RCU_READ_LOCK_GUARD (); -if (count_ignored) { -RAMBLOCK_FOREACH_MIGRATABLE (block) { - total += block->used_length; - } - }else { -RAMBLOCK_FOREACH_NOT_IGNORED (block) { - total += block->used_length; - } +RAMBLOCK_FOREACH_MIGRATABLE (block) { + total += block->used_length; }return total; }uint64_t ram_bytes_total (void ) { -return ram_bytes_total_common (false); +RAMBlock *block; +uint64_t total =0 ; + +RCU_READ_LOCK_GUARD (); + +RAMBLOCK_FOREACH_NOT_IGNORED (block) { + total += block->used_length; + } +return total; }static void xbzrle_load_setup (void )
@@ -3227,7 +3229,8 @@ static int ram_save_setup(QEMUFile *f, void *opaque) (*rsp)->pss[RAM_CHANNEL_PRECOPY ].pss_channel = f; WITH_RCU_READ_LOCK_GUARD() { -qemu_put_be64 (f,ram_bytes_total_common (true) |RAM_SAVE_FLAG_MEM_SIZE ); +qemu_put_be64 (f,ram_bytes_total_with_ignored () + |RAM_SAVE_FLAG_MEM_SIZE );RAMBLOCK_FOREACH_MIGRATABLE (block) {qemu_put_byte (f,strlen (block->idstr));
Signature · 1 line
2.39.1
[PULL 01/17] migration: Remove spurious files
In reply to Juan QuintelaI introduced spurious files on my tree during a rebase:
commit ebfc57871506b3fe36cc41f69ee3ad31a34afd63
Author: Zhenzhong Duan <[email protected]> Date: Mon Oct 17 15:53:51 2022 +0800
multifd: Fix flush of zero copy page send request
Make IO channel flush call after the inflight request has been drained in multifd thread, or else we may missed to flush the inflight request.
To make things worse, it appears like Zhenzhong is the one to blame.
for(int i=0; i < 1000000; i++) {
printf("I will not do rebases when I am tired\n");
}Sorry, Juan.
--- .../x86_64-quintela-devices.mak | 7 - .../x86_64-quintela2-devices.mak | 6 - migration/multifd.c.orig | 1274 ----------------- 3 files changed, 1287 deletions(-) delete mode 100644 configs/devices/x86_64-softmmu/x86_64-quintela-devices.mak delete mode 100644 configs/devices/x86_64-softmmu/x86_64-quintela2-devices.mak delete mode 100644 migration/multifd.c.orig
configs/devices/x86_64-softmmu/x86_64-quintela-devices.mak deleted +0 -7
@@ -1,7 +0,0 @@ -# Boards: -# -CONFIG_ISAPC=n -CONFIG_I440FX=n -CONFIG_Q35=n -CONFIG_MICROVM=y -
configs/devices/x86_64-softmmu/x86_64-quintela2-devices.mak deleted +0 -6
@@ -1,6 +0,0 @@ -# Boards: -# -CONFIG_ISAPC=y -CONFIG_I440FX=y -CONFIG_Q35=y -CONFIG_MICROVM=y
migration/multifd.c.orig deleted +0 -1274
1,274 lines not on this page · raw message
Signature · 1 line
2.39.1
[PULL 04/17] linux-headers: Update to v6.1
In reply to Juan QuintelaFrom: Peter Xu <[email protected]>
--- include/standard-headers/drm/drm_fourcc.h | 34 ++++- include/standard-headers/linux/ethtool.h | 63 +++++++- include/standard-headers/linux/fuse.h | 6 +- .../linux/input-event-codes.h | 1 + include/standard-headers/linux/virtio_blk.h | 19 +++ linux-headers/asm-generic/hugetlb_encode.h | 26 ++-- linux-headers/asm-generic/mman-common.h | 2 + linux-headers/asm-mips/mman.h | 2 + linux-headers/asm-riscv/kvm.h | 4 + linux-headers/linux/kvm.h | 1 + linux-headers/linux/psci.h | 14 ++ linux-headers/linux/userfaultfd.h | 4 + linux-headers/linux/vfio.h | 142 ++++++++++++++++++ 13 files changed, 298 insertions(+), 20 deletions(-)
include/standard-headers/drm/drm_fourcc.h +30 -4
@@ -98,18 +98,42 @@ extern "C" {#define DRM_FORMAT_INVALID 0/* color index */ +#define DRM_FORMAT_C1 fourcc_code('C', '1', ' ', ' ')/* [7:0] C0:C1:C2:C3:C4:C5:C6:C7 1:1:1:1:1:1:1:1 eight pixels/byte */ +#define DRM_FORMAT_C2 fourcc_code('C', '2', ' ', ' ')/* [7:0] C0:C1:C2:C3 2:2:2:2 four pixels/byte */ +#define DRM_FORMAT_C4 fourcc_code('C', '4', ' ', ' ')/* [7:0] C0:C1 4:4 two pixels/byte */ #define DRM_FORMAT_C8 fourcc_code('C', '8', ' ', ' ')/* [7:0] C */ -/* 8 bpp Red */ +/* 1 bpp Darkness (inverse relationship between channel value and brightness) */ +#define DRM_FORMAT_D1 fourcc_code('D', '1', ' ', ' ')/* [7:0] D0:D1:D2:D3:D4:D5:D6:D7 1:1:1:1:1:1:1:1 eight pixels/byte */ + +/* 2 bpp Darkness (inverse relationship between channel value and brightness) */ +#define DRM_FORMAT_D2 fourcc_code('D', '2', ' ', ' ')/* [7:0] D0:D1:D2:D3 2:2:2:2 four pixels/byte */ + +/* 4 bpp Darkness (inverse relationship between channel value and brightness) */ +#define DRM_FORMAT_D4 fourcc_code('D', '4', ' ', ' ')/* [7:0] D0:D1 4:4 two pixels/byte */ + +/* 8 bpp Darkness (inverse relationship between channel value and brightness) */ +#define DRM_FORMAT_D8 fourcc_code('D', '8', ' ', ' ')/* [7:0] D */ + +/* 1 bpp Red (direct relationship between channel value and brightness) */ +#define DRM_FORMAT_R1 fourcc_code('R', '1', ' ', ' ')/* [7:0] R0:R1:R2:R3:R4:R5:R6:R7 1:1:1:1:1:1:1:1 eight pixels/byte */ + +/* 2 bpp Red (direct relationship between channel value and brightness) */ +#define DRM_FORMAT_R2 fourcc_code('R', '2', ' ', ' ')/* [7:0] R0:R1:R2:R3 2:2:2:2 four pixels/byte */ + +/* 4 bpp Red (direct relationship between channel value and brightness) */ +#define DRM_FORMAT_R4 fourcc_code('R', '4', ' ', ' ')/* [7:0] R0:R1 4:4 two pixels/byte */ + +/* 8 bpp Red (direct relationship between channel value and brightness) */ #define DRM_FORMAT_R8 fourcc_code('R', '8', ' ', ' ')/* [7:0] R */ -/* 10 bpp Red */ +/* 10 bpp Red (direct relationship between channel value and brightness) */ #define DRM_FORMAT_R10 fourcc_code('R', '1', '0', ' ')/* [15:0] x:R 6:10 little endian */ -/* 12 bpp Red */ +/* 12 bpp Red (direct relationship between channel value and brightness) */ #define DRM_FORMAT_R12 fourcc_code('R', '1', '2', ' ')/* [15:0] x:R 4:12 little endian */ -/* 16 bpp Red */ +/* 16 bpp Red (direct relationship between channel value and brightness) */ #define DRM_FORMAT_R16 fourcc_code('R', '1', '6', ' ')/* [15:0] R little endian */ /* 16 bpp RG */
@@ -204,7 +228,9 @@ extern "C" {#define DRM_FORMAT_VYUY fourcc_code('V', 'Y', 'U', 'Y')/* [31:0] Y1:Cb0:Y0:Cr0 8:8:8:8 little endian */ #define DRM_FORMAT_AYUV fourcc_code('A', 'Y', 'U', 'V')/* [31:0] A:Y:Cb:Cr 8:8:8:8 little endian */ +#define DRM_FORMAT_AVUY8888 fourcc_code('A', 'V', 'U', 'Y')/* [31:0] A:Cr:Cb:Y 8:8:8:8 little endian */ #define DRM_FORMAT_XYUV8888 fourcc_code('X', 'Y', 'U', 'V')/* [31:0] X:Y:Cb:Cr 8:8:8:8 little endian */ +#define DRM_FORMAT_XVUY8888 fourcc_code('X', 'V', 'U', 'Y')/* [31:0] X:Cr:Cb:Y 8:8:8:8 little endian */ #define DRM_FORMAT_VUY888 fourcc_code('V', 'U', '2', '4')/* [23:0] Cr:Cb:Y 8:8:8 little endian */ #define DRM_FORMAT_VUY101010 fourcc_code('V', 'U', '3', '0')/* Y followed by U then V, 10:10:10. Non-linear modifier only */
include/standard-headers/linux/ethtool.h +61 -2
88 lines not on this page · raw message
include/standard-headers/linux/fuse.h +5 -1
@@ -194,6 +194,9 @@* - add FUSE_SECURITY_CTX init flag * - add security context to create, mkdir, symlink, and mknod requests * - add FUSE_HAS_INODE_DAX, FUSE_ATTR_DAX +* +* 7.37 +* - add FUSE_TMPFILE */ #ifndef _LINUX_FUSE_H
@@ -225,7 +228,7 @@#define FUSE_KERNEL_VERSION 7/** Minor version number of this interface */ -#define FUSE_KERNEL_MINOR_VERSION 36 +#define FUSE_KERNEL_MINOR_VERSION 37/** The node ID of the root inode */ #define FUSE_ROOT_ID 1
@@ -533,6 +536,7 @@ enum fuse_opcode {FUSE_SETUPMAPPING =48 ,FUSE_REMOVEMAPPING =49 ,FUSE_SYNCFS =50 , +FUSE_TMPFILE =51 ,/* CUSE specific operations */ CUSE_INIT = 4096,
include/standard-headers/linux/input-event-codes.h +1 -0
@@ -862,6 +862,7 @@#define ABS_TOOL_WIDTH 0x1c#define ABS_VOLUME 0x20 +#define ABS_PROFILE 0x21#define ABS_MISC 0x28
include/standard-headers/linux/virtio_blk.h +19 -0
@@ -40,6 +40,7 @@#define VIRTIO_BLK_F_MQ 12/* support more than one vq */ #define VIRTIO_BLK_F_DISCARD 13/* DISCARD is supported */ #define VIRTIO_BLK_F_WRITE_ZEROES 14/* WRITE ZEROES is supported */ +#define VIRTIO_BLK_F_SECURE_ERASE 16/* Secure Erase is supported */ /* Legacy feature bits */ #ifndef VIRTIO_BLK_NO_LEGACY
@@ -119,6 +120,21 @@ struct virtio_blk_config {uint8_t write_zeroes_may_unmap;uint8_t unused1[3 ]; + +/* the next 3 entries are guarded by VIRTIO_BLK_F_SECURE_ERASE */ +/* +* The maximum secure erase sectors (in 512-byte sectors) for +* one segment. +*/ +__virtio32 max_secure_erase_sectors; +/* +* The maximum number of secure erase segments in a +* secure erase command. +*/ +__virtio32 max_secure_erase_seg; +/* Secure erase commands must be aligned to this number of sectors. */ +__virtio32 secure_erase_sector_alignment; + }QEMU_PACKED ;/*
@@ -153,6 +169,9 @@ struct virtio_blk_config {/* Write zeroes command */ #define VIRTIO_BLK_T_WRITE_ZEROES 13 +/* Secure erase command */ +#define VIRTIO_BLK_T_SECURE_ERASE 14 +#ifndef VIRTIO_BLK_NO_LEGACY /* Barrier before this op. */ #define VIRTIO_BLK_T_BARRIER 0x80000000
linux-headers/asm-generic/hugetlb_encode.h +13 -13
@@ -20,18 +20,18 @@#define HUGETLB_FLAG_ENCODE_SHIFT 26#define HUGETLB_FLAG_ENCODE_MASK 0x3f -#define HUGETLB_FLAG_ENCODE_16KB (14 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_64KB (16 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_512KB (19 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_1MB (20 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_2MB (21 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_8MB (23 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_16MB (24 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_32MB (25 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_256MB (28 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_512MB (29 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_1GB (30 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_2GB (31 << HUGETLB_FLAG_ENCODE_SHIFT) -#define HUGETLB_FLAG_ENCODE_16GB (34 << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_16KB (14U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_64KB (16U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_512KB (19U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_1MB (20U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_2MB (21U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_8MB (23U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_16MB (24U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_32MB (25U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_256MB (28U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_512MB (29U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_1GB (30U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_2GB (31U << HUGETLB_FLAG_ENCODE_SHIFT) +#define HUGETLB_FLAG_ENCODE_16GB (34U << HUGETLB_FLAG_ENCODE_SHIFT)#endif /* _ASM_GENERIC_HUGETLB_ENCODE_H_ */
linux-headers/asm-generic/mman-common.h +2 -0
@@ -77,6 +77,8 @@#define MADV_DONTNEED_LOCKED 24/* like DONTNEED, but drop locked pages too */ +#define MADV_COLLAPSE 25/* Synchronous hugepage collapse */ +/* compatibility flags */ #define MAP_FILE 0
linux-headers/asm-mips/mman.h +2 -0
@@ -103,6 +103,8 @@#define MADV_DONTNEED_LOCKED 24/* like DONTNEED, but drop locked pages too */ +#define MADV_COLLAPSE 25/* Synchronous hugepage collapse */ +/* compatibility flags */ #define MAP_FILE 0
linux-headers/asm-riscv/kvm.h +4 -0
@@ -48,6 +48,7 @@ struct kvm_sregs {/* CONFIG registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */ struct kvm_riscv_config {unsigned long isa; +unsigned long zicbom_block_size; };/* CORE registers for KVM_GET_ONE_REG and KVM_SET_ONE_REG */
@@ -98,6 +99,9 @@ enum KVM_RISCV_ISA_EXT_ID {KVM_RISCV_ISA_EXT_M ,KVM_RISCV_ISA_EXT_SVPBMT ,KVM_RISCV_ISA_EXT_SSTC , +KVM_RISCV_ISA_EXT_SVINVAL , +KVM_RISCV_ISA_EXT_ZIHINTPAUSE , +KVM_RISCV_ISA_EXT_ZICBOM , KVM_RISCV_ISA_EXT_MAX, };
linux-headers/linux/kvm.h +1 -0
@@ -1175,6 +1175,7 @@ struct kvm_ppc_resize_hpt {#define KVM_CAP_VM_DISABLE_NX_HUGE_PAGES 220#define KVM_CAP_S390_ZPCI_OP 221#define KVM_CAP_S390_CPU_TOPOLOGY 222 +#define KVM_CAP_DIRTY_LOG_RING_ACQ_REL 223#ifdef KVM_CAP_IRQ_ROUTING
linux-headers/linux/psci.h +14 -0
@@ -48,12 +48,26 @@#define PSCI_0_2_FN64_MIGRATE_INFO_UP_CPU PSCI_0_2_FN64(7)#define PSCI_1_0_FN_PSCI_FEATURES PSCI_0_2_FN(10) +#define PSCI_1_0_FN_CPU_FREEZE PSCI_0_2_FN(11) +#define PSCI_1_0_FN_CPU_DEFAULT_SUSPEND PSCI_0_2_FN(12) +#define PSCI_1_0_FN_NODE_HW_STATE PSCI_0_2_FN(13)#define PSCI_1_0_FN_SYSTEM_SUSPEND PSCI_0_2_FN(14)#define PSCI_1_0_FN_SET_SUSPEND_MODE PSCI_0_2_FN(15) +#define PSCI_1_0_FN_STAT_RESIDENCY PSCI_0_2_FN(16) +#define PSCI_1_0_FN_STAT_COUNT PSCI_0_2_FN(17) +#define PSCI_1_1_FN_SYSTEM_RESET2 PSCI_0_2_FN(18) +#define PSCI_1_1_FN_MEM_PROTECT PSCI_0_2_FN(19) +#define PSCI_1_1_FN_MEM_PROTECT_CHECK_RANGE PSCI_0_2_FN(19) +#define PSCI_1_0_FN64_CPU_DEFAULT_SUSPEND PSCI_0_2_FN64(12) +#define PSCI_1_0_FN64_NODE_HW_STATE PSCI_0_2_FN64(13)#define PSCI_1_0_FN64_SYSTEM_SUSPEND PSCI_0_2_FN64(14) +#define PSCI_1_0_FN64_STAT_RESIDENCY PSCI_0_2_FN64(16) +#define PSCI_1_0_FN64_STAT_COUNT PSCI_0_2_FN64(17) +#define PSCI_1_1_FN64_SYSTEM_RESET2 PSCI_0_2_FN64(18) +#define PSCI_1_1_FN64_MEM_PROTECT_CHECK_RANGE PSCI_0_2_FN64(19)/* PSCI v0.2 power state encoding for CPU_SUSPEND function */ #define PSCI_0_2_POWER_STATE_ID_MASK 0xffff
linux-headers/linux/userfaultfd.h +4 -0
@@ -12,6 +12,10 @@#include <linux/types.h> +/* ioctls for /dev/userfaultfd */ +#define USERFAULTFD_IOC 0xAA +#define USERFAULTFD_IOC_NEW _IO(USERFAULTFD_IOC, 0x00) +/* * If the UFFDIO_API is upgraded someday, the UFFDIO_UNREGISTER and * UFFDIO_WAKE ioctls should be defined as _IOW and not as _IOR. In
linux-headers/linux/vfio.h +142 -0
148 lines not on this page · raw message
Signature · 1 line
2.39.1
[PULL 09/17] migration: Calculate ram size once
In reply to Juan QuintelaWe are recalculating ram size continously, when we know that it don't change during migration. Create a field in RAMState to track it.
--- migration/ram.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
migration/ram.c +5 -2
@@ -330,6 +330,8 @@ struct RAMState {PageSearchStatus pss[RAM_CHANNEL_MAX ];/* UFFD file descriptor, used in 'write-tracking' migration */ int uffdio_fd; +/* total ram size in bytes */ +uint64_t ram_bytes_total;/* Last block that we have visited searching for dirty pages */ RAMBlock *last_seen_block;/* Last dirty target page we have sent */
@@ -2546,7 +2548,7 @@ static int ram_find_and_save_block(RAMState *rs)int pages =0 ;/* No dirty page as there is zero RAM */ -if (!ram_bytes_total ()) { +if (!rs->ram_bytes_total) {return pages; }
@@ -3009,13 +3011,14 @@ static int ram_state_init(RAMState **rsp)qemu_mutex_init (&(*rsp)->bitmap_mutex);qemu_mutex_init (&(*rsp)->src_page_req_mutex);QSIMPLEQ_INIT (&(*rsp)->src_page_requests); + (*rsp)->ram_bytes_total =ram_bytes_total ();/* * Count the total number of pages used by ram blocks not including any * gaps due to alignment or unplugs. * This must match with the initial values of dirty bitmap. */ - (*rsp)->migration_dirty_pages =ram_bytes_total () >>TARGET_PAGE_BITS ; + (*rsp)->migration_dirty_pages = (*rsp)->ram_bytes_total >>TARGET_PAGE_BITS ;ram_state_reset (*rsp);return 0 ;
Signature · 1 line
2.39.1
[PULL 17/17] migration: Postpone postcopy preempt channel to be after main
In reply to Juan QuintelaFrom: Peter Xu <[email protected]>
Postcopy with preempt-mode enabled needs two channels to communicate. The order of channel establishment is not guaranteed. It can happen that the dest QEMU got the preempt channel connection request before the main channel is established, then the migration may make no progress even during precopy due to the wrong order.
To fix it, create the preempt channel only if we know the main channel is established.
For a general postcopy migration, we delay it until postcopy_start(), that's where we already went through some part of precopy on the main channel. To make sure dest QEMU has already established the channel, we wait until we got the first PONG received. That's something we do at the start of precopy when postcopy enabled so it's guaranteed to happen sooner or later.
For a postcopy recovery, we delay it to qemu_savevm_state_resume_prepare() where we'll have round trips of data on bitmap synchronizations, which means the main channel must have been established.
--- migration/migration.h | 6 ++++ migration/postcopy-ram.h | 2 +- migration/migration.c | 72 ++++++++++++++++++++++++++++++---------- migration/postcopy-ram.c | 17 ++++++++-- migration/savevm.c | 6 +++- 5 files changed, 82 insertions(+), 21 deletions(-)
migration/migration.h +6 -0
@@ -116,6 +116,12 @@ struct MigrationIncomingState {unsigned int postcopy_channels;/* QEMUFile for postcopy only; it'll be handled by a separate thread */ QEMUFile *postcopy_qemufile_dst; +/* +* When postcopy_qemufile_dst is properly setup, this sem is posted. +* One can wait on this semaphore to wait until the preempt channel is +* properly setup. +*/ +QemuSemaphore postcopy_qemufile_dst_done;/* Postcopy priority thread is used to receive postcopy requested pages */ QemuThread postcopy_prio_thread;bool postcopy_prio_thread_created;
migration/postcopy-ram.h +1 -1
@@ -192,6 +192,6 @@ enum PostcopyChannels {void postcopy_preempt_new_channel (MigrationIncomingState *mis,QEMUFile *file);void postcopy_preempt_setup (MigrationState *s); -int postcopy_preempt_wait_channel (MigrationState *s); +int postcopy_preempt_establish_channel (MigrationState *s);#endif
migration/migration.c +55 -17
120 lines not on this page · raw message
migration/postcopy-ram.c +15 -2
@@ -1197,6 +1197,11 @@ int postcopy_ram_incoming_setup(MigrationIncomingState *mis) }if (migrate_postcopy_preempt ()) { +/* +* The preempt channel is established in asynchronous way. Wait +* for its completion. +*/ +qemu_sem_wait (&mis->postcopy_qemufile_dst_done);/* * This thread needs to be created after the temp pages because * it'll fetch RAM_CHANNEL_POSTCOPY PostcopyTmpPage immediately.
@@ -1544,6 +1549,7 @@ void postcopy_preempt_new_channel(MigrationIncomingState *mis, QEMUFile *file)*/ qemu_file_set_blocking (file, true); mis->postcopy_qemufile_dst = file; +qemu_sem_post (&mis->postcopy_qemufile_dst_done);trace_postcopy_preempt_new_channel (); }
@@ -1612,14 +1618,21 @@ out:postcopy_preempt_send_channel_done (s, ioc, local_err); } -/* Returns 0 if channel established, -1 for error. */ -int postcopy_preempt_wait_channel (MigrationState *s) +/* +* This function will kick off an async task to establish the preempt +* channel, and wait until the connection setup completed. Returns 0 if +* channel established, -1 for error. +*/ +int postcopy_preempt_establish_channel (MigrationState *s) {/* If preempt not enabled, no need to wait */ if (!migrate_postcopy_preempt ()) {return 0 ; } +/* Kick off async task to establish preempt channel */ +postcopy_preempt_setup (s); +/* * We need the postcopy preempt channel to be established before * starting doing anything.
migration/savevm.c +5 -1
@@ -2200,7 +2200,11 @@ static int loadvm_postcopy_handle_resume(MigrationIncomingState *mis)qemu_sem_post (&mis->postcopy_pause_sem_fault);if (migrate_postcopy_preempt ()) { -/* The channel should already be setup again; make sure of it */ +/* +* The preempt channel will be created in async manner, now let's +* wait for it and make sure it's created. +*/ +qemu_sem_wait (&mis->postcopy_qemufile_dst_done);assert (mis->postcopy_qemufile_dst);/* Kick the fast ram load thread too */ qemu_sem_post (&mis->postcopy_pause_sem_fast_load);
Signature · 1 line
2.39.1
[PULL 10/17] migration: Make ram_save_target_page() a pointer
In reply to Juan QuintelaWe are going to create a new function for multifd latest in the series.
--- migration/ram.c | 19 +++++++++++++++---- 1 file changed, 15 insertions(+), 4 deletions(-)
migration/ram.c +15 -4
@@ -452,6 +452,13 @@ void dirty_sync_missed_zero_copy(void) ram_counters.dirty_sync_missed_zero_copy++; } +struct MigrationOps { +int (*ram_save_target_page)(RAMState *rs,PageSearchStatus *pss); +}; +typedef struct MigrationOps MigrationOps ; + +MigrationOps *migration_ops; +CompressionStats compression_counters;struct CompressParam {
@@ -2295,14 +2302,14 @@ static bool save_compress_page(RAMState *rs, PageSearchStatus *pss, }/** -* ram_save_target_page: save one target page +* ram_save_target_page_legacy: save one target page * * Returns the number of pages written * * @rs: current RAM state * @pss: data about the page we want to send */ -static int ram_save_target_page (RAMState *rs,PageSearchStatus *pss) +static int ram_save_target_page_legacy (RAMState *rs,PageSearchStatus *pss) {RAMBlock *block = pss->block;ram_addr_t offset = ((ram_addr_t )pss->page) <<TARGET_PAGE_BITS ;
@@ -2428,7 +2435,7 @@ static int ram_save_host_page_urgent(PageSearchStatus *pss)if (page_dirty) {/* Be strict to return code; it must be 1, or what else? */ -if (ram_save_target_page (rs, pss) !=1 ) { +if (migration_ops->ram_save_target_page (rs, pss) !=1 ) {error_report_once ("%s: ram_save_target_page failed" , __func__); ret =-1 ; goto out;
@@ -2497,7 +2504,7 @@ static int ram_save_host_page(RAMState *rs, PageSearchStatus *pss)if (preempt_active) {qemu_mutex_unlock (&rs->bitmap_mutex); } - tmppages =ram_save_target_page (rs, pss); + tmppages = migration_ops->ram_save_target_page (rs, pss);if (tmppages >=0 ) { pages += tmppages;/*
@@ -2697,6 +2704,8 @@ static void ram_save_cleanup(void *opaque)xbzrle_cleanup ();compress_threads_save_cleanup ();ram_state_cleanup (rsp); +g_free (migration_ops); + migration_ops =NULL ; }static void ram_state_reset (RAMState *rs)
@@ -3252,6 +3261,8 @@ static int ram_save_setup(QEMUFile *f, void *opaque)ram_control_before_iterate (f,RAM_CONTROL_SETUP );ram_control_after_iterate (f,RAM_CONTROL_SETUP ); + migration_ops =g_malloc0 (sizeof (MigrationOps)); + migration_ops->ram_save_target_page = ram_save_target_page_legacy; ret =multifd_send_sync_main (f);if (ret <0 ) {return ret;
Signature · 1 line
2.39.1
[PULL 12/17] AVX512 support for xbzrle_encode_buffer
In reply to Juan QuintelaFrom: ling xu <[email protected]>
This commit is the same with [PATCH v6 1/2], and provides avx512 support for xbzrle_encode_buffer function to accelerate xbzrle encoding speed. Runtime check of avx512
support and benchmark for this feature are added. Compared with C
version of xbzrle_encode_buffer function, avx512 version can achieve
50%-70% performance improvement on benchmarking. In addition, if dirty
data is randomly located in 4K page, the avx512 version can achieve
almost 140% performance gain.
--- meson.build | 17 +++++ migration/xbzrle.h | 4 ++ migration/ram.c | 38 +++++++++-- migration/xbzrle.c | 124 ++++++++++++++++++++++++++++++++++ meson_options.txt | 2 + scripts/meson-buildoptions.sh | 3 + 6 files changed, 184 insertions(+), 4 deletions(-)
meson.build +17 -0
@@ -2351,6 +2351,22 @@ config_host_data.set('CONFIG_AVX512F_OPT', get_option('avx512f') \ int main(int argc, char *argv[]) { return bar(argv[argc - 1]); } '''), error_message: 'AVX512F not available').allowed()) +config_host_data.set('CONFIG_AVX512BW_OPT', get_option('avx512bw') \ + .require(have_cpuid_h, error_message: 'cpuid.h not available, cannot enable AVX512BW') \ + .require(cc.links(''' + #pragma GCC push_options + #pragma GCC target("avx512bw") + #include <cpuid.h> + #include <immintrin.h> + static int bar(void *a) { + + __m512i *x = a; + __m512i res= _mm512_abs_epi8(*x); + return res[1]; + } + int main(int argc, char *argv[]) { return bar(argv[0]); } + '''), error_message: 'AVX512BW not available').allowed()) + have_pvrdma = get_option('pvrdma') \ .require(rdma.found(), error_message: 'PVRDMA requires OpenFabrics libraries') \ .require(cc.compiles(gnu_source_prefix + '''
@@ -3783,6 +3799,7 @@ summary_info += {'debug stack usage': get_option('debug_stack_usage')} summary_info += {'mutex debugging': get_option('debug_mutex')} summary_info += {'memory allocator': get_option('malloc')} summary_info += {'avx2 optimization': config_host_data.get('CONFIG_AVX2_OPT')} +summary_info += {'avx512bw optimization': config_host_data.get('CONFIG_AVX512BW_OPT')} summary_info += {'avx512f optimization': config_host_data.get('CONFIG_AVX512F_OPT')} summary_info += {'gprof enabled': get_option('gprof')} summary_info += {'gcov': get_option('b_coverage')}
migration/xbzrle.h +4 -0
@@ -18,4 +18,8 @@ int xbzrle_encode_buffer(uint8_t *old_buf, uint8_t *new_buf, int slen, uint8_t *dst, int dlen);int xbzrle_decode_buffer (uint8_t *src,int slen,uint8_t *dst,int dlen); +#if defined(CONFIG_AVX512BW_OPT ) +int xbzrle_encode_buffer_avx512 (uint8_t *old_buf,uint8_t *new_buf,int slen, +uint8_t *dst,int dlen); +#endif #endif
migration/ram.c +34 -4
@@ -83,6 +83,35 @@/* 0x80 is reserved in migration.h start with 0x100 next */ #define RAM_SAVE_FLAG_COMPRESS_PAGE 0x100 +int (*xbzrle_encode_buffer_func)(uint8_t *,uint8_t *,int , +uint8_t *,int ) = xbzrle_encode_buffer; +#if defined(CONFIG_AVX512BW_OPT ) +#include "qemu/cpuid.h" +static void __attribute__((constructor))init_cpu_flag (void ) +{ +unsigned max =__get_cpuid_max (0 ,NULL ); +int a, b, c, d; +if (max >=1 ) { +__cpuid (1 , a, b, c, d); +/* We must check that AVX is not just available, but usable. */ +if ((c & bit_OSXSAVE) && (c & bit_AVX) && max >=7 ) { +int bv; + __asm("xgetbv" :"=a" (bv),"=d" (d) :"c" (0 )); +__cpuid_count (7 ,0 , a, b, c, d); +/* +* 0xe6: +* XCR0[7:5] = 111b (OPMASK state, upper 256-bit of ZMM0-ZMM15 +* and ZMM16-ZMM31 state are enabled by OS) +* XCR0[2:1] = 11b (XMM state and YMM state are enabled by OS) +*/ +if ((bv &0xe6 ) ==0xe6 && (b & bit_AVX512BW)) { + xbzrle_encode_buffer_func = xbzrle_encode_buffer_avx512; + } + } + } +} +#endif +XBZRLECacheStats xbzrle_counters;/* used by the search for pages to send */
@@ -806,10 +835,11 @@ static int save_xbzrle_page(RAMState *rs, PageSearchStatus *pss,memcpy (XBZRLE .current_buf, *current_data,TARGET_PAGE_SIZE );/* XBZRLE encoding (if there is no overflow) */ - encoded_len =xbzrle_encode_buffer (prev_cached_page,XBZRLE .current_buf, -TARGET_PAGE_SIZE ,XBZRLE .encoded_buf, -TARGET_PAGE_SIZE ); - + encoded_len =xbzrle_encode_buffer_func (prev_cached_page, +XBZRLE .current_buf, +TARGET_PAGE_SIZE , +XBZRLE .encoded_buf, +TARGET_PAGE_SIZE );/* * Update the cache contents, so that it corresponds to the data * sent, in all cases except where we skip the page.
migration/xbzrle.c +124 -0
127 lines not on this page · raw message
meson_options.txt +2 -0
@@ -104,6 +104,8 @@ option('avx2', type: 'feature', value: 'auto', description: 'AVX2 optimizations') option('avx512f', type: 'feature', value: 'disabled', description: 'AVX512F optimizations') +option('avx512bw', type: 'feature', value: 'auto', + description: 'AVX512BW optimizations') option('keyring', type: 'feature', value: 'auto', description: 'Linux keyring support')
scripts/meson-buildoptions.sh +3 -0
@@ -70,6 +70,7 @@ meson_options_help() { printf "%s\n" ' attr attr/xattr support' printf "%s\n" ' auth-pam PAM access control' printf "%s\n" ' avx2 AVX2 optimizations' + printf "%s\n" ' avx512bw AVX512BW optimizations' printf "%s\n" ' avx512f AVX512F optimizations' printf "%s\n" ' blkio libblkio block device driver' printf "%s\n" ' bochs bochs image format support'
@@ -198,6 +199,8 @@ _meson_option_parse() { --disable-auth-pam) printf "%s" -Dauth_pam=disabled ;; --enable-avx2) printf "%s" -Davx2=enabled ;; --disable-avx2) printf "%s" -Davx2=disabled ;; + --enable-avx512bw) printf "%s" -Davx512bw=enabled ;; + --disable-avx512bw) printf "%s" -Davx512bw=disabled ;; --enable-avx512f) printf "%s" -Davx512f=enabled ;; --disable-avx512f) printf "%s" -Davx512f=disabled ;; --enable-gcov) printf "%s" -Db_coverage=true ;;
Signature · 1 line
2.39.1
Re: [PULL 12/17] AVX512 support for xbzrle_encode_buffer
Quoted 4 lines · Juan Quintela
From: ling xu <[email protected]>
This commit is the same with [PATCH v6 1/2], and provides avx512 support for
xbzrle_encode_bufferfunction to accelerate xbzrle encoding speed.
Hi!
Just a hint for future patches: Please don't put version information into the patch description itself, this "PATCH v6 1/2" will now end up in the git history without its context.
Information like that should go into cover letters or below the "---" marker only.
Thanks,
Thomas
[PULL 14/17] migration: Rework multi-channel checks on URI
In reply to Juan QuintelaFrom: Peter Xu <[email protected]>
The whole idea of multi-channel checks was not properly done, IMHO.
Currently we check multi-channel in a lot of places, but actually that's not needed because we only need to check it right after we get the URI and that should be it.
If the URI check succeeded, we should never need to check it again because we must have it. If it check fails, we should fail immediately on either the qmp_migrate or qmp_migrate_incoming, instead of failingg it later after the connection established.
Neither should we fail any set capabiliities like what we used to do here:
5ad15e8614 ("migration: allow enabling mutilfd for specific protocol only", 2021-10-19)
Because logically the URI will only be set later after the capability is set, so it doesn't make a lot of sense to check the URI type when setting the capability, because we're checking the cap with an old URI passed in, and that may not even be the URI we're going to use later.
This patch mostly reverted all such checks for before, dropping the variable migrate_allow_multi_channels and helpers. Instead, add a common helper to check URI for multi-channels for either qmp_migrate and qmp_migrate_incoming and that should do all the proper checks. The failure will only trigger with the "migrate" or "migrate_incoming" command, or when user specified "-incoming xxx" where "xxx" is not "defer".
--- migration/migration.h | 3 --- migration/migration.c | 49 +++++++++++++++++++++------------------- migration/multifd.c | 12 ++-------- migration/postcopy-ram.c | 6 ----- 4 files changed, 28 insertions(+), 42 deletions(-)
migration/migration.h +0 -3
@@ -474,7 +474,4 @@ void migration_cancel(const Error *error);void populate_vfio_info (MigrationInfo *info);void postcopy_temp_page_reset (PostcopyTmpPage *tmp_page); -bool migrate_multi_channels_is_allowed (void ); -void migrate_protocol_allow_multi_channels (bool allow); -#endif
migration/migration.c +26 -23
97 lines not on this page · raw message
migration/multifd.c +2 -10
@@ -516,7 +516,7 @@ void multifd_save_cleanup(void) {int i; -if (!migrate_use_multifd () || !migrate_multi_channels_is_allowed ()) { +if (!migrate_use_multifd ()) {return ; }multifd_send_terminate_threads (NULL );
@@ -913,10 +913,6 @@ int multifd_save_setup(Error **errp)if (!migrate_use_multifd ()) {return 0 ; } -if (!migrate_multi_channels_is_allowed ()) { -error_setg (errp,"multifd is not supported by current protocol" ); -return -1 ; - } thread_count =migrate_multifd_channels (); multifd_send_state =g_malloc0 (sizeof (*multifd_send_state));
@@ -1021,7 +1017,7 @@ int multifd_load_cleanup(Error **errp) {int i; -if (!migrate_use_multifd () || !migrate_multi_channels_is_allowed ()) { +if (!migrate_use_multifd ()) {return 0 ; }multifd_recv_terminate_threads (NULL );
@@ -1172,10 +1168,6 @@ int multifd_load_setup(Error **errp)return 0 ; } -if (!migrate_multi_channels_is_allowed ()) { -error_setg (errp,"multifd is not supported by current protocol" ); -return -1 ; - } thread_count =migrate_multifd_channels (); multifd_recv_state =g_malloc0 (sizeof (*multifd_recv_state)); multifd_recv_state->params =g_new0 (MultiFDRecvParams, thread_count);
migration/postcopy-ram.c +0 -6
@@ -1635,12 +1635,6 @@ int postcopy_preempt_setup(MigrationState *s, Error **errp)return 0 ; } -if (!migrate_multi_channels_is_allowed ()) { -error_setg (errp,"Postcopy preempt is not supported as current " -"migration stream does not support multi-channels." ); -return -1 ; - } -/* Kick an async task to connect */ socket_send_channel_create (postcopy_preempt_send_channel_new, s);
Signature · 1 line
2.39.1
[PULL 11/17] migration: I messed state_pending_exact/estimate
In reply to Juan QuintelaI called the helper function from the wrong top level function.
This code was introduced in:
commit c8df4a7aeffcb46020f610526eea621fa5b0cd47
Author: Juan Quintela <[email protected]> Date: Mon Oct 3 02:00:03 2022 +0200
migration: Split save_live_pending() into state_pending_*
We split the function into to:
- state_pending_estimate: We estimate the remaining state size without stopping the machine.
- state pending_exact: We calculate the exact amount of remaining state.
Thanks to Avihai Horon <[email protected]> for finding it.
When we introduced that patch, we enden calling
state_pending_estimate() helper from qemu_savevm_statepending_exact() andstate_pending_exact() helper from qemu_savevm_statepending_estimate()
This patch fixes it.
--- migration/savevm.c | 50 +++++++++++++++++++++++----------------------- 1 file changed, 25 insertions(+), 25 deletions(-)
migration/savevm.c +25 -25
@@ -1551,31 +1551,6 @@ void qemu_savevm_state_pending_estimate(uint64_t *res_precopy_only, *res_compatible =0 ; *res_postcopy_only =0 ; - QTAILQ_FOREACH(se, &savevm_state.handlers, entry) { -if (!se->ops || !se->ops->state_pending_exact) { -continue ; - } -if (se->ops->is_active) { -if (!se->ops->is_active (se->opaque)) { -continue ; - } - } - se->ops->state_pending_exact (se->opaque, - res_precopy_only, res_compatible, - res_postcopy_only); - } -} - -void qemu_savevm_state_pending_exact (uint64_t *res_precopy_only, -uint64_t *res_compatible, -uint64_t *res_postcopy_only) -{ -SaveStateEntry *se; - - *res_precopy_only =0 ; - *res_compatible =0 ; - *res_postcopy_only =0 ; - QTAILQ_FOREACH(se, &savevm_state.handlers, entry) {if (!se->ops || !se->ops->state_pending_estimate) {continue ;
@@ -1591,6 +1566,31 @@ void qemu_savevm_state_pending_exact(uint64_t *res_precopy_only, } } +void qemu_savevm_state_pending_exact (uint64_t *res_precopy_only, +uint64_t *res_compatible, +uint64_t *res_postcopy_only) +{ +SaveStateEntry *se; + + *res_precopy_only =0 ; + *res_compatible =0 ; + *res_postcopy_only =0 ; + + QTAILQ_FOREACH(se, &savevm_state.handlers, entry) { +if (!se->ops || !se->ops->state_pending_exact) { +continue ; + } +if (se->ops->is_active) { +if (!se->ops->is_active (se->opaque)) { +continue ; + } + } + se->ops->state_pending_exact (se->opaque, + res_precopy_only, res_compatible, + res_postcopy_only); + } +} +void qemu_savevm_state_cleanup (void ) {SaveStateEntry *se;
Signature · 1 line
2.39.1
[PULL 16/17] migration: Add a semaphore to count PONGs
In reply to Juan QuintelaFrom: Peter Xu <[email protected]>
This is mostly useless, but useful for us to know whether the main channel is correctly established without changing the migration protocol.
--- migration/migration.h | 6 ++++++ migration/migration.c | 3 +++ 2 files changed, 9 insertions(+)
migration/migration.h +6 -0
@@ -276,6 +276,12 @@ struct MigrationState {*/ bool rp_thread_created;QemuSemaphore rp_sem; +/* +* We post to this when we got one PONG from dest. So far it's an +* easy way to know the main channel has successfully established +* on dest QEMU. +*/ +QemuSemaphore rp_pong_acks; } rp_state;double mbps;
migration/migration.c +3 -0
@@ -3025,6 +3025,7 @@ retry:case MIG_RP_MSG_PONG : tmp32 =ldl_be_p (buf);trace_source_return_path_thread_pong (tmp32); +qemu_sem_post (&ms->rp_state.rp_pong_acks);break ;case MIG_RP_MSG_REQ_PAGES :
@@ -4524,6 +4525,7 @@ static void migration_instance_finalize(Object *obj)qemu_sem_destroy (&ms->postcopy_pause_sem);qemu_sem_destroy (&ms->postcopy_pause_rp_sem);qemu_sem_destroy (&ms->rp_state.rp_sem); +qemu_sem_destroy (&ms->rp_state.rp_pong_acks);qemu_sem_destroy (&ms->postcopy_qemufile_src_sem);error_free (ms->error); }
@@ -4570,6 +4572,7 @@ static void migration_instance_init(Object *obj)qemu_sem_init (&ms->postcopy_pause_sem,0 );qemu_sem_init (&ms->postcopy_pause_rp_sem,0 );qemu_sem_init (&ms->rp_state.rp_sem,0 ); +qemu_sem_init (&ms->rp_state.rp_pong_acks,0 );qemu_sem_init (&ms->rate_limit_sem,0 );qemu_sem_init (&ms->wait_unplug_sem,0 );qemu_sem_init (&ms->postcopy_qemufile_src_sem,0 );
Signature · 1 line
2.39.1
[PULL 13/17] Update bench-code for addressing CI problem
In reply to Juan QuintelaFrom: ling xu <[email protected]>
Unit test code is in test-xbzrle.c, and benchmark code is in xbzrle-bench.c for performance benchmarking. we have modified xbzrle-bench.c to address CI problem.
--- MAINTAINERS | 1 + tests/bench/xbzrle-bench.c | 473 +++++++++++++++++++++++++++++++++++++ tests/unit/test-xbzrle.c | 42 +++- tests/bench/meson.build | 4 + 4 files changed, 514 insertions(+), 6 deletions(-) create mode 100644 tests/bench/xbzrle-bench.c
MAINTAINERS +1 -0
@@ -3100,6 +3100,7 @@ F: include/migration/ F: include/qemu/userfaultfd.h F: migration/ F: scripts/vmstate-static-checker.py +F: tests/bench/xbzrle-bench.c F: tests/vmstate-static-checker-data/ F: tests/qtest/migration-test.c F: docs/devel/migration.rst
tests/bench/xbzrle-bench.c new file +473 -0
473 lines not on this page · raw message
tests/unit/test-xbzrle.c +36 -6
78 lines not on this page · raw message
tests/bench/meson.build +4 -0
@@ -3,6 +3,10 @@ qht_bench = executable('qht-bench', sources: 'qht-bench.c', dependencies: [qemuutil]) +xbzrle_bench = executable('xbzrle-bench', + sources: 'xbzrle-bench.c', + dependencies: [qemuutil,migration]) + executable('atomic_add-bench', sources: files('atomic_add-bench.c'), dependencies: [qemuutil],
Signature · 1 line
2.39.1
[PULL 07/17] migration: Make find_dirty_block() return a single parameter
In reply to Juan QuintelaWe used to return two bools, just return a single int with the following meaning:
old return / again / new return
false false PAGE_ALL_CLEAN false true PAGE_TRY_AGAIN true true PAGE_DIRTY_FOUND /* We don't care about again at all */
--- migration/ram.c | 37 ++++++++++++++++++++++--------------- 1 file changed, 22 insertions(+), 15 deletions(-)
migration/ram.c +22 -15
@@ -1546,17 +1546,23 @@ retry:return pages; } +#define PAGE_ALL_CLEAN 0 +#define PAGE_TRY_AGAIN 1 +#define PAGE_DIRTY_FOUND 2/** * find_dirty_block: find the next dirty page and update any state * associated with the search process. * -* Returns true if a page is found +* Returns: +* PAGE_ALL_CLEAN: no dirty page found, give up +* PAGE_TRY_AGAIN: no dirty page found, retry for next block +* PAGE_DIRTY_FOUND: dirty page found * * @rs: current RAM state * @pss: data about the state of the current dirty page scan * @again: set to false if the search has scanned the whole of RAM */ -static bool find_dirty_block (RAMState *rs,PageSearchStatus *pss,bool *again) +static int find_dirty_block (RAMState *rs,PageSearchStatus *pss) {/* Update pss->page for the next dirty bit in ramblock */ pss_find_next_dirty (pss);
@@ -1567,8 +1573,7 @@ static bool find_dirty_block(RAMState *rs, PageSearchStatus *pss, bool *again)* We've been once around the RAM and haven't found anything. * Give up. */ - *again = false; -return false; +return PAGE_ALL_CLEAN ; }if (!offset_in_ramblock (pss->block, ((ram_addr_t )pss->page) <<TARGET_PAGE_BITS )) {
@@ -1597,13 +1602,10 @@ static bool find_dirty_block(RAMState *rs, PageSearchStatus *pss, bool *again) } }/* Didn't find anything this time, but try again on the new block */ - *again = true; -return false; +return PAGE_TRY_AGAIN ; }else { -/* Can go around again, but... */ - *again = true; -/* We've found something so probably don't need to */ -return true; +/* We've found something */ +return PAGE_DIRTY_FOUND ; } }
@@ -2562,18 +2564,23 @@ static int ram_find_and_save_block(RAMState *rs)pss_init (pss, rs->last_seen_block, rs->last_page); -do { +while (true) {if (!get_queued_page (rs, pss)) {/* priority queue empty, so just search for something dirty */ -bool again = true; -if (!find_dirty_block (rs, pss, &again)) { -if (!again) { +int res =find_dirty_block (rs, pss); +if (res !=PAGE_DIRTY_FOUND ) { +if (res ==PAGE_ALL_CLEAN ) {break ; + }else if (res ==PAGE_TRY_AGAIN ) { +continue ; } } } pages =ram_save_host_page (rs, pss); - }while (!pages); +if (pages) { +break ; + } + } rs->last_seen_block = pss->block; rs->last_page = pss->page;
Signature · 1 line
2.39.1
[PULL 15/17] migration: Cleanup postcopy_preempt_setup()
In reply to Juan QuintelaFrom: Peter Xu <[email protected]>
Since we just dropped the only case where postcopy_preempt_setup() can return an error, it doesn't need a retval anymore because it never fails. Move the preempt check to the caller, preparing it to be used elsewhere to do nothing but as simple as kicking the async connection.
--- migration/postcopy-ram.h | 2 +- migration/migration.c | 8 ++------ migration/postcopy-ram.c | 8 +------- 3 files changed, 4 insertions(+), 14 deletions(-)
migration/postcopy-ram.h +1 -1
@@ -191,7 +191,7 @@ enum PostcopyChannels { };void postcopy_preempt_new_channel (MigrationIncomingState *mis,QEMUFile *file); -int postcopy_preempt_setup (MigrationState *s,Error **errp); +void postcopy_preempt_setup (MigrationState *s);int postcopy_preempt_wait_channel (MigrationState *s);#endif
migration/migration.c +2 -6
@@ -4347,12 +4347,8 @@ void migrate_fd_connect(MigrationState *s, Error *error_in) }/* This needs to be done before resuming a postcopy */ -if (postcopy_preempt_setup (s, &local_err)) { -error_report_err (local_err); -migrate_set_state (&s->state,MIGRATION_STATUS_SETUP , -MIGRATION_STATUS_FAILED ); -migrate_fd_cleanup (s); -return ; +if (migrate_postcopy_preempt ()) { +postcopy_preempt_setup (s); }if (resume) {
migration/postcopy-ram.c +1 -7
@@ -1629,16 +1629,10 @@ int postcopy_preempt_wait_channel(MigrationState *s)return s->postcopy_qemufile_src ?0 :-1 ; } -int postcopy_preempt_setup (MigrationState *s,Error **errp) +void postcopy_preempt_setup (MigrationState *s) { -if (!migrate_postcopy_preempt ()) { -return 0 ; - } -/* Kick an async task to connect */ socket_send_channel_create (postcopy_preempt_send_channel_new, s); - -return 0 ; }static void postcopy_pause_ram_fast_load (MigrationIncomingState *mis)
Signature · 1 line
2.39.1
[PULL 06/17] migration: Simplify ram_find_and_save_block()
In reply to Juan QuintelaWe will need later that find_dirty_block() return errors, so simplify the loop.
--- migration/ram.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-)
migration/ram.c +9 -11
@@ -2542,7 +2542,6 @@ static int ram_find_and_save_block(RAMState *rs) {PageSearchStatus *pss = &rs->pss[RAM_CHANNEL_PRECOPY ];int pages =0 ; -bool again, found;/* No dirty page as there is zero RAM */ if (!ram_bytes_total ()) {
@@ -2564,18 +2563,17 @@ static int ram_find_and_save_block(RAMState *rs)pss_init (pss, rs->last_seen_block, rs->last_page);do { - again = true; - found =get_queued_page (rs, pss); - -if (!found) { +if (!get_queued_page (rs, pss)) {/* priority queue empty, so just search for something dirty */ - found =find_dirty_block (rs, pss, &again); +bool again = true; +if (!find_dirty_block (rs, pss, &again)) { +if (!again) { +break ; + } + } } - -if (found) { - pages =ram_save_host_page (rs, pss); - } - }while (!pages && again); + pages =ram_save_host_page (rs, pss); + }while (!pages); rs->last_seen_block = pss->block; rs->last_page = pss->page;
Signature · 1 line
2.39.1
[PULL 05/17] util/userfaultfd: Support /dev/userfaultfd
In reply to Juan QuintelaFrom: Peter Xu <[email protected]>
Teach QEMU to use /dev/userfaultfd when it existed and fallback to the system call if either it's not there or doesn't have enough permission.
Firstly, as long as the app has permission to access /dev/userfaultfd, it always have the ability to trap kernel faults which QEMU mostly wants. Meanwhile, in some context (e.g. containers) the userfaultfd syscall can be forbidden, so it can be the major way to use postcopy in a restricted environment with strict seccomp setup.
--- util/userfaultfd.c | 32 ++++++++++++++++++++++++++++++++ util/trace-events | 1 + 2 files changed, 33 insertions(+)
util/userfaultfd.c +32 -0
@@ -18,10 +18,42 @@#include <poll.h> #include <sys/syscall.h> #include <sys/ioctl.h> +#include <fcntl.h> + +typedef enum { +UFFD_UNINITIALIZED =0 , +UFFD_USE_DEV_PATH , +UFFD_USE_SYSCALL , +}uffd_open_mode ;int uffd_open (int flags) {#if defined(__NR_userfaultfd) +static uffd_open_mode open_mode; +static int uffd_dev; + +/* Detect how to generate uffd desc when run the 1st time */ +if (open_mode ==UFFD_UNINITIALIZED ) { +/* +* Make /dev/userfaultfd the default approach because it has better +* permission controls, meanwhile allows kernel faults without any +* privilege requirement (e.g. SYS_CAP_PTRACE). +*/ + uffd_dev =open ("/dev/userfaultfd" ,O_RDWR |O_CLOEXEC ); +if (uffd_dev >=0 ) { + open_mode =UFFD_USE_DEV_PATH ; + }else { +/* Fallback to the system call */ + open_mode =UFFD_USE_SYSCALL ; + } +trace_uffd_detect_open_mode (open_mode); + } + +if (open_mode ==UFFD_USE_DEV_PATH ) { +assert (uffd_dev >=0 ); +return ioctl (uffd_dev,USERFAULTFD_IOC_NEW , flags); + } +return syscall (__NR_userfaultfd, flags);#else return -EINVAL ;
util/trace-events +1 -0
@@ -93,6 +93,7 @@ qemu_vfio_region_info(const char *desc, uint64_t region_ofs, uint64_t region_siz qemu_vfio_pci_map_bar(int index, uint64_t region_ofs, uint64_t region_size, int ofs, void *host) "map region bar#%d addr 0x%"PRIx64" size 0x%"PRIx64" ofs 0x%x host %p" #userfaultfd.c +uffd_detect_open_mode(int mode) "%d" uffd_query_features_nosys(int err) "errno: %i" uffd_query_features_api_failed(int err) "errno: %i" uffd_create_fd_nosys(int err) "errno: %i"
Signature · 1 line
2.39.1
Quoted 29 lines · Juan Quintela
The following changes since commit
417296c8d8588f782018d01a317f88957e9786d6:tests/qtest/netdev-socket: Raise connection timeout to 60 seconds (2023-02-09 11:23:53 +0000)
are available in the Git repository at:
https://gitlab.com/juan.quintela/qemu.git tags/migration-20230209-pull-request
for you to fetch changes up to
858191aebda251a4d1e3bc77b238096673241cdd:migration: Postpone postcopy preempt channel to be after main (2023-02-09 21:26:02 +0100)
----------------------------------------------------------------
Migration Pull requestHi
This are all the reviewed patches for migration:
- AVX512 support for xbzrle (Ling Xu)
-/dev/userfaultdsupport (Peter Xu)
- Improve ordering of channels (Peter Xu)
- multifd cleanups (Li Zhang)
- Remove spurious files from last merge (me) Rebase makes that to you
- Fix mixup between state_pending_{exact,estimate} (me)
- Cache RAM size during migration (me)
- cleanup several functions (me)
Quoted 3 lines
Please apply.
----------------------------------------------------------------
Fails to build the user-mode emulators:
https://gitlab.com/qemu-project/qemu/-/jobs/3749435025
In file included from ../authz/base.c:24:../authz/trace.h:1:10: fatal error: trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"
https://gitlab.com/qemu-project/qemu/-/jobs/3749435094
In file included from ../authz/simple.c:23:../authz/trace.h:1:10: fatal error: trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"
https://gitlab.com/qemu-project/qemu/-/jobs/3749434963
In file included from ../authz/listfile.c:23:../authz/trace.h:1:10: fatal error: trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"
etc
thanks
-- PMM
Quoted 30 lines · Peter Maydell
Quoted 29 lines · Juan Quintela
The following changes since commit
417296c8d8588f782018d01a317f88957e9786d6:tests/qtest/netdev-socket: Raise connection timeout to 60 seconds (2023-02-09 11:23:53 +0000)
are available in the Git repository at:
https://gitlab.com/juan.quintela/qemu.git tags/migration-20230209-pull-request
for you to fetch changes up to
858191aebda251a4d1e3bc77b238096673241cdd:migration: Postpone postcopy preempt channel to be after main (2023-02-09 21:26:02 +0100)
----------------------------------------------------------------
Migration Pull requestHi
This are all the reviewed patches for migration:
- AVX512 support for xbzrle (Ling Xu)
-/dev/userfaultdsupport (Peter Xu)
- Improve ordering of channels (Peter Xu)
- multifd cleanups (Li Zhang)
- Remove spurious files from last merge (me) Rebase makes that to you
- Fix mixup between state_pending_{exact,estimate} (me)
- Cache RAM size during migration (me)
- cleanup several functions (me)
Quoted 5 lines
Quoted 3 lines
Please apply.
----------------------------------------------------------------
Fails to build the user-mode emulators:
This is weird.
$ ls *user -d aarch64_be-linux-user m68k-linux-user ppc-linux-user aarch64-linux-user microblazeel-linux-user riscv32-linux-user alpha-linux-user microblaze-linux-user riscv64-linux-user armeb-linux-user mips64el-linux-user s390x-linux-user arm-linux-user mips64-linux-user sh4eb-linux-user bsd-user mipsel-linux-user sh4-linux-user common-user mips-linux-user sparc32plus-linux-user cris-linux-user mipsn32el-linux-user sparc64-linux-user hexagon-linux-user mipsn32-linux-user sparc-linux-user hppa-linux-user nios2-linux-user x86_64-linux-user i386-linux-user or1k-linux-user xtensaeb-linux-user linux-user ppc64le-linux-user xtensa-linux-user loongarch64-linux-user ppc64-linux-user
I think I build all user emulators locally.
Quoted 6 lines
https://gitlab.com/qemu-project/qemu/-/jobs/3749435025
In file included from
../authz/base.c:24:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"
This series only have one change for traces:
util/trace-events +1 -0
@@ -93,6 +93,7 @@ qemu_vfio_region_info(const char *desc, uint64_t region_ofs, uint64_t region_siz qemu_vfio_pci_map_bar(int index, uint64_t region_ofs, uint64_t region_size, int ofs, void *host) "map region bar#%d addr 0x%"PRIx64" size 0x%"PRIx64" ofs 0x%x host %p" #userfaultfd.c +uffd_detect_open_mode(int mode) "%d" uffd_query_features_nosys(int err) "errno: %i" uffd_query_features_api_failed(int err) "errno: %i" uffd_create_fd_nosys(int err) "errno: %i"
Rest of trace mentions are for the removal of migration.multifd.c.orig
And I don't play with authentication at all.
This is Fedora 37.
Quoted 4 lines
https://gitlab.com/qemu-project/qemu/-/jobs/3749435094
In file included from../authz/simple.c:23:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directory
Problem is that this trace file is not generated, but I can think how any change that I did can influence this.
Quoted 8 lines
1 | #include "
trace/trace-authz.h"https://gitlab.com/qemu-project/qemu/-/jobs/3749434963
In file included from../authz/listfile.c:23:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"
Looking at the ouptut of these, they are not informatives at all.
I am going to try to compile linux-user without system, and see if that brings a clue.
But I don't know what is going on here.
Later, Juan.
Quoted 51 lines · Juan Quintela
Quoted 1 line · Peter Maydell
Fails to build the user-mode emulators:
This is weird.
Quoted 6 lines
https://gitlab.com/qemu-project/qemu/-/jobs/3749435025
In file included from
../authz/base.c:24:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"This series only have one change for traces:
util/trace-events +1
-0@@ -93,6 +93,7 @@ qemu_vfio_region_info(const char *desc, uint64_t region_ofs, uint64_t region_siz qemu_vfio_pci_map_bar(int index, uint64_t region_ofs, uint64_t region_size, int ofs, void *host) "map region bar#%d addr 0x%"PRIx64" size 0x%"PRIx64" ofs 0x%x host %p" #userfaultfd.c +uffd_detect_open_mode(int mode) "%d" uffd_query_features_nosys(int err) "errno: %i" uffd_query_features_api_failed(int err) "errno: %i" uffd_create_fd_nosys(int err) "errno: %i"Rest of trace mentions are for the removal of migration.multifd.c.orig
And I don't play with authentication at all.
This is Fedora 37.
Quoted 4 lines
https://gitlab.com/qemu-project/qemu/-/jobs/3749435094
In file included from../authz/simple.c:23:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directoryProblem is that this trace file is not generated, but I can think how any change that I did can influence this.
Quoted 8 lines
1 | #include "
trace/trace-authz.h"https://gitlab.com/qemu-project/qemu/-/jobs/3749434963
In file included from../authz/listfile.c:23:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"
Quoted 4 lines
Looking at the ouptut of these, they are not informatives at all.
I am going to try to compile linux-user without system, and see if that brings a clue.
Yes, I suspect this is a "user-mode only build" specific failure (you may need --disable-system --disable-tools to see it).
meson.build only puts authz into trace_events_subdirs "if have_block" (which is to say "if have_system or have_tools"). However the bit of meson.build that says "subdir('authz') does not have the same condition on it -- it's just been put in the list without any condition on it. So I think that in a build-only-user-emulators config meson will not generate trace events for the subdirectory but will try to build it, which falls over.
Contrast 'block', 'nbd', 'scsi', which are all guarded by 'if have_block' for their subdir() lines, to match the guard on the trace_events_subdirs. OTOH 'io' is also mismatched-guards...
Why this only shows up with your pullreq I have no idea.
thanks
-- PMM
Quoted 54 lines · Peter Maydell
Quoted 53 lines · Juan Quintela
Quoted 1 line · Peter Maydell
Fails to build the user-mode emulators:
This is weird.
Quoted 6 lines
https://gitlab.com/qemu-project/qemu/-/jobs/3749435025
In file included from
../authz/base.c:24:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"This series only have one change for traces:
util/trace-events +0
-0@@ -93,6 +93,7 @@ qemu_vfio_region_info(const char *desc, uint64_t region_ofs, uint64_t region_siz qemu_vfio_pci_map_bar(int index, uint64_t region_ofs, uint64_t
region_size, int ofs, void *host) "map region bar#%d addr 0x%"PRIx64" size 0x%"PRIx64" ofs 0x%x host %p"#userfaultfd.c
+uffd_detect_open_mode(int mode) "%d"
uffd_query_features_nosys(int err)"errno: %i"uffd_query_features_api_failed(int err)"errno: %i"uffd_create_fd_nosys(int err)"errno: %i"Rest of trace mentions are for the removal of migration.multifd.c.orig
And I don't play with authentication at all.
This is Fedora 37.
Quoted 4 lines
https://gitlab.com/qemu-project/qemu/-/jobs/3749435094
In file included from../authz/simple.c:23:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directoryProblem is that this trace file is not generated, but I can think how any change that I did can influence this.
Quoted 8 lines
1 | #include "
trace/trace-authz.h"https://gitlab.com/qemu-project/qemu/-/jobs/3749434963
In file included from../authz/listfile.c:23:../authz/trace.h:1:10: fatal error:trace/trace-authz.h: No such file or directory
1 | #include "trace/trace-authz.h"
Quoted 7 lines
Quoted 4 lines
Looking at the ouptut of these, they are not informatives at all.
I am going to try to compile linux-user without system, and see if that brings a clue.
Yes, I suspect this is a "user-mode only build" specific failure (you may need --disable-system --disable-tools to see it).
git-bisect is my friend O:-)
And yes, the problem was in my PULL request.
Again, I don't know why it fails.
tests/bench/meson.build +2 -0
@@ -3,9 +3,11 @@ qht_bench = executable('qht-bench', sources: 'qht-bench.c', dependencies: [qemuutil]) +if have_system xbzrle_bench = executable('xbzrle-bench', sources: 'xbzrle-bench.c', dependencies: [qemuutil,migration]) +endif executable('atomic_add-bench', sources: files('atomic_add-bench.c'),
This make it works.
And no, I still not have a single clue how creating a new executable in tests/bench/ can make trace files not to be generated somewhere else.
Quoted 8 lines
meson.build only puts authz into
trace_events_subdirs"ifhave_block" (which is to say "ifhave_systemorhave_tools"). However the bit of meson.build that says "subdir('authz') does not have the same condition on it -- it's just been put in the list without any condition on it. So I think that in a build-only-user-emulators config meson will not generate trace events for the subdirectory but will try to build it, which falls over.
Quoted 5 lines
Contrast 'block', 'nbd', 'scsi', which are all guarded by 'if
have_block' for theirsubdir()lines, to match the guard on thetrace_events_subdirs. OTOH 'io' is also mismatched-guards...Why this only shows up with your pullreq I have no idea.
xbzrle_bench.
Notice that dependency on migration.
As said, I compile every user and system targets that compile in fedora before I submit a PULL request.
But it is getting clear than that is not enough anymore.
Sorry about the noise.
Will resubmit the PULL requset.
Later, Juan.
Quoted 19 lines · Juan Quintela
Again, I don't know why it fails.
tests/bench/meson.build +2
-0@@ -3,9 +3,11 @@ qht_bench = executable('qht-bench', sources: 'qht-bench.c', dependencies: [qemuutil]) +if have_system xbzrle_bench = executable('xbzrle-bench', sources: 'xbzrle-bench.c', dependencies: [qemuutil,migration]) +endif executable('atomic_add-bench', sources: files('atomic_add-bench.c'),
Quoted 1 line
This make it works.
Before you added your test, meson had no need to compile any of the object files in 'migration', so it didn't. Now you tell meson to build a new executable, and it says "OK, I must build these object files". Only it turns out that they won't actually compile in this config, so you get an error.
The same issue can happen in good old Make :-)
thanks
-- PMM