to #29722367

On some specific hardware with 128 bytes LLC cacheline, tk_core may
cause false sharing problem. We can align it to 128 bytes so that
it won't be affected by other global variables.

This change will make a bit waste on cache utilization but get good
number of performance improvement. So for both 64 and 128 bytes aligned
LLC cacheline, we adjust tk_core memory layout to avoid potential cacheline
contention.
Signed-off-by: NPeng Wang <rocking@linux.alibaba.com>
Acked-by: NShanpei Chen <shanpeic@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>

fix #30070241

Enable the vsyscall emulate help to support some legency docker images
wich is based on CentOS 6, etc.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

fix #30012285

Open the UIO Kconfig for x86_64.
Signed-off-by: NErwei Deng <erwei@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

to #29613419

To be consistent, disable low limit in arm as it is not used, also
enable io latency in x86.
Signed-off-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Acked-by: NXiaoguang Wang <xiaoguang.wang@linux.alibaba.com>

task #28910367
These two are added to alinux recently, update configs to reflect
them.
Signed-off-by: NLiu Bo <bo.liu@linux.alibaba.com>
Reviewed-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

task #29355864

Disable the following builtin modules, which is needless for modern
kernel:
 - CONFIG_VIRTIO_BLK_SCSI=y
 - CONFIG_SYSCTL_SYSCALL=y
 - CONFIG_ISA_BUS=y
 - CONFIG_NET_VENDOR_CADENCE=y
 - CONFIG_NET_VENDOR_CORTINA=y
 - CONFIG_NET_VENDOR_I825XX=y
 - CONFIG_NET_VENDOR_NETERION=y
 - CONFIG_NET_VENDOR_NI=y
 - CONFIG_NET_VENDOR_PACKET_ENGINES=y
 - CONFIG_NET_VENDOR_SOCIONEXT=y
 - CONFIG_XFS_RT=y
 - CONFIG_DEBUG_SG=y
 - CONFIG_DEBUG_NOTIFIERS=y
 - CONFIG_DEBUG_CREDENTIALS=y
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Acked-by: NDust Li <dust.li@linux.alibaba.com>

fix #29334855

Intel RDT and AMD QoS are used to cache monitoring/allocation and memory
bandwidth monitoring and allocation. In order to enabled Intel RDT and
AMD QoS, CONFIG_RESCTRL has to be configured as Y.
Signed-off-by: NZelin Deng <zelin.deng@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

to #27804112
Signed-off-by: NXuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: NDust Li <dust.li@linux.alibaba.com>
Reviewed-by: NYa Zhao <zhaoya123@linux.alibaba.com>
Reviewed-by: NCambda Zhu <cambda@linux.alibaba.com>

fix #29335392

DPC stands for downstream port containment, it will halt PCIE traffic
below a downstream port after an unmasked uncorrectable error is
detected at or below the port, so that it can avoid the potential spread
of any data corruption.
Signed-off-by: NZelin Deng <zelin.deng@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

to #29272054

AF_XDP is a new AF family that support usespace applications
communicate with XDP program directly.
One promising use case is UDP

Both x86_64 and aarch64 are enabled
Signed-off-by: NDust Li <dust.li@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

fix #29180329

CONFIG_REFCOUNT_FULL is used for debugging mainly,
for release kernel, it's better to diable it.
This patch disables both x86 and aarch64 for release
kernel.

This has a pretty large performance penalty for
will-it-scale:signal1_process when the process
number are large.
Signed-off-by: NDust Li <dust.li@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #27305291

Enabled the following configs for NVDIMM support:
- CONFIG_ACPI_NFIT=m
- CONFIG_NVDIMM_KEYS=y
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Reviewed-by: NYang Shi <yang.shi@linux.alibaba.com>

to #27182371

Restored all the tuned configs for Cloud Kernel before, to keep
the unified configs for both x86_64 and ARM64.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

to #27182371

Reconfig the ARM64 with Alibaba internal kernel help to keep
the unified kernel configs.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

task #25552995

add PSEUDO_NMI configuration item
Signed-off-by: NZou Cao <zoucao@linux.alibaba.com>
Reviewed-by: Nluanshi <zhangliguang@linux.alibaba.com>

task #28924046

This patch enables extra hardware errata for
ARM64 N1 platform and just for aarch64, not for x86.
Signed-off-by: NBin Yu <jkchen@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

fix #28925962

In faas's VPC network mode, we should enable this feature to support
multi-tenancy's snat on a node.
Signed-off-by: NZhiyuan Hou <zhiyuan2048@linux.alibaba.com>
Acked-by: NDust Li <dust.li@linux.alibaba.com>

fix #28886284

On AMD platforms cpu frequency was not able to be tuned as there's no
cpufreq driver registered -- intel_pstate has been enabled but it only
can be loaded on Intel CPUs. Hence after evaluated and validated on AMD
platforms, we decide to enable acpi-cpufreq.
acpi-cpufreq won't impact on intel_pstate on Intel platforms as
intel_pstate will be loaded in device_initcall while acpi-cpufreq will
be loaded in late_initcall. This sequence ensure intel_pstate can be
loaded but acpi-cpufreq can not on Intel platforms.
Signed-off-by: NZelin Deng <zelin.deng@linux.alibaba.com>
Reviewed-by: NCaspar Zhang <caspar@linux.alibaba.com>
Reviewed-by: NArtie Ding <fulin.dn@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

fix #28198752

ext4 encryption will increase lock contention when opening directory and
result in performance drop in case will-it-scale open1.
Since we don't have explicit usecases as of now, so we decide to disabed
it.
Signed-off-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Reviewed-by: NDust Li <dust.li@linux.alibaba.com>

fix #28506983

Some ARM machines may have large memory capacity (e.g., more than 256G),
or large hole(s) in memory layout among nodes.

Kernel with CONFIG_ARM64_VA_BITS as 39 has the linear region size as
256G, and the memory that we will not be able to cover with the linear
mapping shall be removed. This may cause part of the physical memory to
become unavailable, system deadlock on memory, or even boot failure, on
such ARM machines.

This changes CONFIG_ARM64_VA_BITS to 48 which supports 128T linear
mapping, in order to adapt to most scenarios.
Signed-off-by: NXu Yu <xuyu@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Reviewed-by: NCaspar Zhang <caspar@linux.alibaba.com>

to #28170604

Or we will see below compile errors:
In file included from net/atm/ioctl.c:24:
./include/net/compat.h:45:40: error: unknown type name 'compat_uptr_t'; did you mean 'compat_time_t'?
   45 |    struct sockaddr __user **save_addr, compat_uptr_t *ptr,
      |                                        ^~~~~~~~~~~~~
      |                                        compat_time_t
./include/net/compat.h:46:4: error: unknown type name 'compat_size_t'; did you mean 'compat_time_t'?
   46 |    compat_size_t *len);
      |    ^~~~~~~~~~~~~
      |    compat_time_t
Signed-off-by: NXiaoguang Wang <xiaoguang.wang@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

fix #27365977

It enables AMD_IOMMU for AMD platforms which support X2APIC.
Because AMD IOMMU driver initializes OMMU protection domains as
direct map, IOMMU_DEFAULT_PASSTHROUGH is needed.
Signed-off-by: NArtie Ding <artie.ding@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

to #27793353

The following configs are set to 'm' to make x86 the same
as aarch64.
  CONFIG_NET_EMATCH_CMP=m
  CONFIG_NET_EMATCH_NBYTE=m
  CONFIG_NET_EMATCH_U32=m
  CONFIG_NET_EMATCH_META=m
  CONFIG_NET_EMATCH_TEXT=m

Some commonly used cases need CONFIG_NET_EMATCH_CMP,
for example:

tc qdisc add dev eth0 root handle 1: prio bands 4
tc qdisc add dev eth0 parent 1:4 handle 40: netem delay 20ms 2ms
tc filter add dev eth0 parent 1: protocol ip prio 4 basic match
			       "cmp(u16 at 2 layer transport eq 3306)
                            and cmp(u8 at 16 layer network eq 10)
                            and cmp(u8 at 17 layer network eq 0)
                            and cmp(u8 at 18 layer network eq 200)
                            and cmp(u8 at 19 layer network eq 45)" flowid 1:4
Signed-off-by: NDust Li <dust.li@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #27778669

Aligned NET_SCHED configs with aarch64, except:

1. CONFIG_NET_SCH_ATM is not enabled since we don't use
   ATM on cloud, and CONFIG_ATM is not enabled
2. CONFIG_NET_SCH_DEFAULT is not set since we still use
   pfifo_fast as the default scheduler.
3. CONFIG_NET_SCH_FQ_CODEL set to 'm' since we don't use
   fq_codel as the default qdisc
Signed-off-by: NDust Li <dust.li@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #27182371

1. build mouse driver as module;
2. disable RT_GROUP_SCHED;
3. set HZ=250.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Suggested-by: NShanpei Chen <shanpeic@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #26353046
Acked-by: NDust Li <dust.li@linux.alibaba.com>
Signed-off-by: Nxuanzhuo <xuanzhuo@linux.alibaba.com>

fix #27497636

Enable ip route multipath and dm multipath, for consistent with arm and
physical kconfig.
Signed-off-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Acked-by: NDust Li <dust.li@linux.alibaba.com>

to #26809468, #25931767

Disable deferred struct page init by default since some stability issue
observed.

A new kernel parameter was planed to enable it for large size instance
on demand, for memory init speed up.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #27362006

Especially, we eanble the following for daishu, which are consistent
with centos8:
CONFIG_SQUASHFS_FILE_DIRECT
CONFIG_SQUASHFS_DECOMP_MULTI_PERCPU
Signed-off-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

fix #27138800
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #26613714
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #26832542
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NCaspar Zhang <caspar@linux.alibaba.com>

to #26648913
Signed-off-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Acked-by: NXiaoguang Wang <xiaoguang.wang@linux.alibaba.com>

to #26581627

Enable a set of config related to kernel trust, including
vptm, IMA, EVM, smack LSM, and SM3, SM4, in which SM algorithm
is compiled into module.

These configs also support the x64 arm platform.

The following are some performance data. The output of
`systemd-analyze` on the startup time of an ECS (executed
three times):

before:
725ms (kernel) + 471ms (initrd) + 3.375s (userspace) = 4.572s
794ms (kernel) + 426ms (initrd) + 3.281s (userspace) = 4.501s
797ms (kernel) + 464ms (initrd) + 3.275s (userspace) = 4.538s

after:
777ms (kernel) + 439ms (initrd) + 3.456s (userspace) = 4.672s
785ms (kernel) + 450ms (initrd) + 3.313s (userspace) = 4.549s
741ms (kernel) + 481ms (initrd) + 3.274s (userspace) = 4.497s
Signed-off-by: NTianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Jia Zhang <zhang.jia@linux.alibaba.com>
Reviewed-by: NShile Zhang <shile.zhang@linux.alibaba.com>

to #26424323

Enable CONFIG_SCHED_SLI by default.
Signed-off-by: NYihao Wu <wuyihao@linux.alibaba.com>
Acked-by: NShanpei Chen <shanpeic@linux.alibaba.com>

to #26536261

Keep the common configs same between x86_64 and aarch64.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #24582903

Update aarch64 configs since gcc version and more minor changes.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #26616987

Disable CONFIG_NFS_V3_ACL and CONFIG_NFSD_V3_ACL for aarch64,
to be same with x86
Signed-off-by: NChunmei Xu <xuchunmei@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>

to #26424368

Enable CONFIG_MEMSLI by default.
Signed-off-by: NXu Yu <xuyu@linux.alibaba.com>
Reviewed-by: NYang Shi <yang.shi@linux.alibaba.com>
Reviewed-by: NXunlei Pang <xlpang@linux.alibaba.com>

to #26589609

Move the configs in tree for bisect, it's convenient for locating configure
related issue.
Signed-off-by: NShile Zhang <shile.zhang@linux.alibaba.com>
Acked-by: NJoseph Qi <joseph.qi@linux.alibaba.com>