VAS windows can be closed in the hypervisor due to lost credits
when the core is removed and the kernel gets fault for NX
requests on these inactive windows. If the NX requests are
issued on these inactive windows, OS gets page faults and the
paste failure will be returned to the user space. If
The VAS window may not be active if the system looses credits and
the NX generates page fault when it receives request on unmap
paste address.
The kernel handles the fault by remap new paste address if the
window is active again, Otherwise return the paste instruction
failure if the executed
Since the VAS windows belong to the VAS hardware resource, the
hypervisor expects the partition to close them on source partition
and reopen them after the partition migrated on the destination
machine.
This handler is called before pseries_suspend() to close these
windows and again invoked
VAS is a hardware engine stays on the chip. So when the partition
migrates, all VAS windows on the source system have to be closed
and reopen them on the destination after migration.
The kernel has to consider both DLPAR CPU and migration events to
take action on VAS windows. So using
The coprocessor capabilities struct is used to get default and
QoS capabilities from the hypervisor during init, DLPAR event and
migration. So instead of allocating this struct for each event,
define global struct and reuse it which allows the migration code
to avoid adding an error path.
Also
VAS windows can be closed in the hypervisor due to lost credits
when the core is removed and the kernel gets fault for NX
requests on these inactive windows. If the NX requests are
issued on these inactive windows, OS gets page faults and the
paste failure will be returned to the user space. If
Virtual Accelerator Switchboard (VAS) is an engine stays on the
chip. So all windows opened on a specific engine belongs to VAS
the chip. The hypervisor expects the partition to close all
active windows on the sources system and reopen them after
migration on the destination machine.
This patch
pseries supports two types of credits - Default (uses normal priority
FIFO) and Qality of service (QoS uses high priority FIFO). The user
decides the number of QoS credits and sets this value with HMC
interface. The total credits for QoS capabilities can be changed
dynamically with HMC interface
The hypervisor provides the available VAS GZIP capabilities such
as default or QoS window type and the target available credits in
each type. This patch creates sysfs entries and exports the target,
used and the available credits for each feature.
This interface can be used by the user space to
The hypervisor assigns vas credits (windows) for each LPAR based
on the number of cores configured in that system. The OS is
expected to release credits when cores are removed, and may
allocate more when cores are added. So there is a possibility of
using excessive credits (windows) in the LPAR
The paste address mapping is done with mmap() after the window is
opened with ioctl. The partition has to close VAS windows in the
hypervisor if it lost credits due to DLPAR core removal. But the
kernel marks these windows inactive until the previously lost
credits are available later. If the
The VAS window may not be active if the system looses credits and
the NX generates page fault when it receives request on unmap
paste address.
The kernel handles the fault by remap new paste address if the
window is active again, Otherwise return the paste instruction
failure if the executed
The user space opens VAS windows and issues NX requests by pasting
CRB on the corresponding paste address mmap. When the system lost
credits due to core removal, the kernel has to close the window in
the hypervisor and make the window inactive by unmapping this paste
address. Also the OS has to
The kernel sets the VAS window with PID when it is opened in
the hypervisor. During DLPAR operation, windows can be closed and
reopened in the hypervisor when the credit is available. So saves
this PID in pseries_vas_window struct when the window is opened
initially and reuse it later during
nr_total/nr_used_credits provides credits usage to user space
via sysfs and the same interface can be used on PowerNV in
future. Changed with proper naming so that applicable on both
pseries and PowerNV.
Signed-off-by: Haren Myneni
Reviewed-by: Nicholas Piggin
---
PowerPC provides HW compression with NX coprocessor. This feature
is available on both PowerNV and PowerVM and included in Linux.
Since each powerpc chip has one NX coprocessor, the VAS introduces
the concept of windows / credits to manage access to this hardware
resource. On powerVM, these
From: guo...@kernel.org
> Sent: 27 February 2022 16:28
>
> From: Christoph Hellwig
>
> Provide a single common definition for the compat_flock and
> compat_flock64 structures using the same tricks as for the native
> variants. Another extra define is added for the packing required on
> x86.
Hi,
I'm investigating a timekeeping problem on an old PPC-based embedded
device and am curious about the update_persistent_clock() routine.
This device runs a 3.10-era kernel but a similar (deprecated) function
exists in the 5.16 kernels so I'll refer to the latter:
In
From: Christoph Hellwig Sent: Sunday, February 27, 2022 6:31 AM
>
> Pass a bool to pass if swiotlb needs to be enabled based on the
> addressing needs and replace the verbose argument with a set of
> flags, including one to force enable bounce buffering.
>
> Note that this patch removes the
Le 25/02/2022 à 11:24, Mahesh Salgaonkar a écrit :
> nmi_enter()/nmi_exit() touches per cpu variables which can lead to kernel
> crash when invoked during real mode interrupt handling (e.g. early HMI/MCE
> interrupt handler) if percpu allocation comes from vmalloc area.
>
> Early HMI/MCE
Le 27/02/2022 à 15:30, Christoph Hellwig a écrit :
> Pass a bool to pass if swiotlb needs to be enabled based on the
> addressing needs and replace the verbose argument with a set of
> flags, including one to force enable bounce buffering.
>
> Note that this patch removes the possibility to
From: Guo Ren
Adds initial skeletal COMPAT Kbuild (Running 32bit U-mode on
64bit S-mode) support.
- Setup kconfig & dummy functions for compiling.
- Implement compat_start_thread by the way.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Arnd Bergmann
Cc: Palmer Dabbelt
---
From: Guo Ren
Now, you can use native gdb on riscv64 for rv32 app debugging.
$ uname -a
Linux buildroot 5.16.0-rc4-00036-gbef6b82fdf23-dirty #53 SMP Mon Dec 20
23:06:53 CST 2021 riscv64 GNU/Linux
$ cat /proc/cpuinfo
processor : 0
hart: 0
isa : rv64imafdcsuh
mmu
From: Guo Ren
Implement compat_setup_rt_frame for sigcontext save & restore. The
main process is the same with signal, but the rv32 pt_regs' size
is different from rv64's, so we needs convert them.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Palmer Dabbelt
Cc: Arnd Bergmann
From: Guo Ren
Reconstruct __setup_additional_pages() by appending vdso info
pointer argument to meet compat_vdso_info requirement. And change
vm_special_mapping *dm, *cm initialization into static.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Palmer Dabbelt
Cc: Arnd Bergmann
From: Guo Ren
There is no vgettimeofday supported in rv32 that makes simple to
generate rv32 vdso code which only needs riscv64 compiler. Other
architectures need change compiler or -m (machine parameter) to
support vdso32 compiling. If rv32 support vgettimeofday (which
cause C compile) in
From: Guo Ren
Detect hardware COMPAT (32bit U-mode) capability in rv64. If not
support COMPAT mode in hw, compat_elf_check_arch would return
false by compat_binfmt_elf.c
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Cc: Arnd Bergmann
Cc: Christoph Hellwig
---
arch/riscv/include/asm/elf.h |
From: Guo Ren
Implement necessary type and macro for compat elf. See the code
comment for detail.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Arnd Bergmann
---
arch/riscv/include/asm/elf.h | 46 +++-
1 file changed, 45 insertions(+), 1
From: Guo Ren
If the current task is in COMPAT mode, set SR_UXL_32 in status for
returning userspace. We need CONFIG _COMPAT to prevent compiling
errors with rv32 defconfig.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Cc: Arnd Bergmann
Cc: Palmer Dabbelt
---
arch/riscv/kernel/process.c |
From: Guo Ren
Implement the entry of compat_sys_call_table[] in asm. Ref to
riscv-privileged spec 4.1.1 Supervisor Status Register (sstatus):
BIT[32:33] = UXL[1:0]:
- 1:32
- 2:64
- 3:128
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Palmer Dabbelt
Cc: Arnd Bergmann
---
From: Guo Ren
Implement compat sys_call_table and some system call functions:
truncate64, ftruncate64, fallocate, pread64, pwrite64,
sync_file_range, readahead, fadvise64_64 which need argument
translation.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Arnd Bergmann
Cc: Palmer
From: Guo Ren
Make TASK_SIZE from const to dynamic detect TIF_32BIT flag
function. Refer to arm64 to implement DEFAULT_MAP_WINDOW_64 for
efi-stub.
Limit 32-bit compatible process in 0-2GB virtual address range
(which is enough for real scenarios), because it could avoid
address sign extend
From: Guo Ren
Implement riscv asm/compat.h for struct compat_xxx,
is_compat_task, compat_user_regset, regset convert.
The rv64 compat.h has inherited most of the structs
from the generic one.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Cc: Arnd Bergmann
Cc: Palmer Dabbelt
---
From: Guo Ren
Let's follow the origin patch's spirit:
The only difference between rv32_defconfig and defconfig is that
rv32_defconfig has CONFIG_ARCH_RV32I=y.
This is helpful to compare rv64-compat-rv32 v.s. rv32-linux.
Fixes: 1b937e8faa87ccfb ("RISC-V: Add separate defconfig for 32bit
From: Guo Ren
Make "uapi asm unistd.h" could be used for architectures' COMPAT
mode. The __SYSCALL_COMPAT is first used in riscv.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Arnd Bergmann
Reviewed-by: Christoph Hellwig
---
include/uapi/asm-generic/unistd.h | 4 ++--
From: Guo Ren
There are 7 64bit architectures that support Linux COMPAT mode to
run 32bit applications. A lot of definitions are duplicate:
- COMPAT_USER_HZ
- COMPAT_RLIM_INFINITY
- COMPAT_OFF_T_MAX
- __compat_uid_t, __compat_uid_t
- compat_dev_t
- compat_ipc_pid_t
- struct compat_flock
From: Guo Ren
RISC-V doesn't neeed compat_stat, so using __ARCH_WANT_COMPAT_STAT
to exclude unnecessary SYSCALL functions.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Reviewed-by: Arnd Bergmann
Reviewed-by: Christoph Hellwig
Cc: Palmer Dabbelt
---
arch/arm64/include/asm/unistd.h | 1 +
From: Guo Ren
The existing per-arch definitions are pretty much historic cruft.
Move SYSVIPC_COMPAT into init/Kconfig.
Signed-off-by: Guo Ren
Signed-off-by: Guo Ren
Acked-by: Arnd Bergmann
Reviewed-by: Christoph Hellwig
Cc: Palmer Dabbelt
---
arch/arm64/Kconfig | 4
From: Christoph Hellwig
Provide a single common definition for the compat_flock and
compat_flock64 structures using the same tricks as for the native
variants. Another extra define is added for the packing required on
x86.
Signed-off-by: Christoph Hellwig
Signed-off-by: Guo Ren
Reviewed-by:
From: Christoph Hellwig
The F_GETLK64/F_SETLK64/F_SETLKW64 fcntl opcodes are only implemented
for the 32-bit syscall APIs, but are also needed for compat handling
on 64-bit kernels.
Consolidate them in unistd.h instead of definining the internal compat
definitions in compat.h, which is rather
From: Christoph Hellwig
Don't bother to define the symbols empty, just don't use them.
That makes the intent a little more clear.
Remove the unused HAVE_ARCH_STRUCT_FLOCK64 define and merge the
32-bit mips struct flock into the generic one.
Add a new __ARCH_FLOCK_EXTRA_SYSID macro following
From: Guo Ren
Currently, most 64-bit architectures (x86, parisc, powerpc, arm64,
s390, mips, sparc) have supported COMPAT mode. But they all have
history issues and can't use standard linux unistd.h. RISC-V would
be first standard __SYSCALL_COMPAT user of include/uapi/asm-generic
/unistd.h.
The
Use the more specific is_swiotlb_active check instead of checking the
global swiotlb_force variable.
Signed-off-by: Christoph Hellwig
Reviewed-by: Anshuman Khandual
---
kernel/dma/direct.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/kernel/dma/direct.h
gets pulled in by all drivers using the DMA API.
Remove x86 internal variables and unnecessary includes from it.
Signed-off-by: Christoph Hellwig
---
arch/x86/include/asm/dma-mapping.h | 11 ---
arch/x86/include/asm/iommu.h | 2 ++
2 files changed, 2 insertions(+), 11
Allow to pass a remap argument to the swiotlb initialization functions
to handle the Xen/x86 remap case. ARM/ARM64 never did any remapping
from xen_swiotlb_fixup, so we don't even need that quirk.
Signed-off-by: Christoph Hellwig
---
arch/arm/xen/mm.c | 23 +++---
Pass a bool to pass if swiotlb needs to be enabled based on the
addressing needs and replace the verbose argument with a set of
flags, including one to force enable bounce buffering.
Note that this patch removes the possibility to force xen-swiotlb
use using swiotlb=force on the command line on
Power SVM wants to allocate a swiotlb buffer that is not restricted to
low memory for the trusted hypervisor scheme. Consolidate the support
for this into the swiotlb_init interface by adding a new flag.
Signed-off-by: Christoph Hellwig
---
arch/powerpc/include/asm/svm.h | 4
The IOMMU table tries to separate the different IOMMUs into different
backends, but actually requires various cross calls.
Rewrite the code to do the generic swiotlb/swiotlb-xen setup directly
in pci-dma.c and then just call into the IOMMU drivers.
Signed-off-by: Christoph Hellwig
---
Use the generic swiotlb initialization helper instead of open coding it.
Signed-off-by: Christoph Hellwig
---
arch/mips/cavium-octeon/dma-octeon.c | 15 ++-
arch/mips/pci/pci-octeon.c | 2 +-
2 files changed, 3 insertions(+), 14 deletions(-)
diff --git
Let the caller chose a zone to allocate from. This will be used
later on by the xen-swiotlb initialization on arm.
Signed-off-by: Christoph Hellwig
Reviewed-by: Anshuman Khandual
---
arch/x86/pci/sta2x11-fixup.c | 2 +-
include/linux/swiotlb.h | 2 +-
kernel/dma/swiotlb.c | 4
Remove the bogus Xen override that was usually larger than the actual
size and just calculate the value on demand. Note that
swiotlb_max_segment still doesn't make sense as an interface and should
eventually be removed.
Signed-off-by: Christoph Hellwig
Reviewed-by: Anshuman Khandual
---
swiotlb_late_init_with_default_size is an overly verbose name that
doesn't even catch what the function is doing, given that the size is
not just a default but the actual requested size.
Rename it to swiotlb_init_late.
Signed-off-by: Christoph Hellwig
Reviewed-by: Anshuman Khandual
---
Hi all,
this series tries to clean up the swiotlb initialization, including
that of swiotlb-xen. To get there is also removes the x86 iommu table
infrastructure that massively obsfucates the initialization path.
Git tree:
git://git.infradead.org/users/hch/misc.git swiotlb-init-cleanup
If force bouncing is enabled we can't release the buffers.
Signed-off-by: Christoph Hellwig
Reviewed-by: Anshuman Khandual
---
kernel/dma/swiotlb.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c
index f1e7ea160b433..36fbf1181d285 100644
---
54 matches
Mail list logo
