Stefan Binding [Thu, 30 Jun 2022 00:23:22 +0000 (01:23 +0100)]
ALSA: hda: hda_cs_dsp_ctl: Add Library to support CS_DSP ALSA controls
The cs35l41 part contains a DSP which is able to run firmware.
The cs_dsp library can be used to control the DSP.
These controls can be exposed to userspace using ALSA controls.
This library adds apis to be able to interface between
cs_dsp and hda drivers and expose the relevant controls as
ALSA controls.
[ Note: the dependency of CONFIG_SND_HDA_CS_DSP_CONTROLS Kconfig is
corrected. Also, this Kconfig isn't enabled now but will be
actually enabled in a later patch -- tiwai ]
ALSA: usb-audio: Use atomic_try_cmpxchg in ep_state_update
Use atomic_try_cmpxchg instead of atomic_cmpxchg (*ptr, old, new) == old in
ep_state_update. x86 CMPXCHG instruction returns success in ZF flag,
so this change saves a compare after cmpxchg (and related move instruction
in front of cmpxchg).
====================
seg6: fix skb checksum for SRH encapsulation/insertion
The Linux kernel supports Segment Routing Header (SRH)
encapsulation/insertion operations by providing the capability to: i)
encapsulate a packet in an outer IPv6 header with a specified SRH; ii)
insert a specified SRH directly after the IPv6 header of the packet.
Note that the insertion operation is also referred to as 'injection'.
The two operations are respectively supported by seg6_do_srh_encap() and
seg6_do_srh_inline(), which operate on the skb associated to the packet as
needed (e.g. adding the necessary headers and initializing them, while
taking care to recalculate the skb checksum).
seg6_do_srh_encap() and seg6_do_srh_inline() do not initialize the payload
length of the IPv6 header, which is carried out by the caller functions.
However, this approach causes the corruption of the skb checksum which
needs to be updated only after initialization of headers is completed
(thanks to Paolo Abeni for detecting this issue).
The patchset fixes the skb checksum corruption by moving the IPv6 header
payload length initialization from the callers of seg6_do_srh_encap() and
seg6_do_srh_inline() directly into these functions.
This patchset is organized as follows:
- patch 1/3, seg6: fix skb checksum evaluation in SRH
encapsulation/insertion;
(* SRH encapsulation/insertion available since v4.10)
- patch 2/3, seg6: fix skb checksum in SRv6 End.B6 and End.B6.Encaps
behaviors;
(* SRv6 End.B6 and End.B6.Encaps behaviors available since v4.14)
- patch 3/3, seg6: bpf: fix skb checksum in bpf_push_seg6_encap();
(* bpf IPv6 Segment Routing helpers available since v4.18)
Andrea Mayer [Tue, 12 Jul 2022 17:58:37 +0000 (19:58 +0200)]
seg6: bpf: fix skb checksum in bpf_push_seg6_encap()
Both helper functions bpf_lwt_seg6_action() and bpf_lwt_push_encap() use
the bpf_push_seg6_encap() to encapsulate the packet in an IPv6 with Segment
Routing Header (SRH) or insert an SRH between the IPv6 header and the
payload.
To achieve this result, such helper functions rely on bpf_push_seg6_encap()
which, in turn, leverages seg6_do_srh_{encap,inline}() to perform the
required operation (i.e. encap/inline).
This patch removes the initialization of the IPv6 header payload length
from bpf_push_seg6_encap(), as it is now handled properly by
seg6_do_srh_{encap,inline}() to prevent corruption of the skb checksum.
Fixes: fe94cc290f53 ("bpf: Add IPv6 Segment Routing helpers") Signed-off-by: Andrea Mayer <andrea.mayer@uniroma2.it> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Andrea Mayer [Tue, 12 Jul 2022 17:58:36 +0000 (19:58 +0200)]
seg6: fix skb checksum in SRv6 End.B6 and End.B6.Encaps behaviors
The SRv6 End.B6 and End.B6.Encaps behaviors rely on functions
seg6_do_srh_{encap,inline}() to, respectively: i) encapsulate the
packet within an outer IPv6 header with the specified Segment Routing
Header (SRH); ii) insert the specified SRH directly after the IPv6
header of the packet.
This patch removes the initialization of the IPv6 header payload length
from the input_action_end_b6{_encap}() functions, as it is now handled
properly by seg6_do_srh_{encap,inline}() to avoid corruption of the skb
checksum.
Fixes: 140f04c33bbc ("ipv6: sr: implement several seg6local actions") Signed-off-by: Andrea Mayer <andrea.mayer@uniroma2.it> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Andrea Mayer [Tue, 12 Jul 2022 17:58:35 +0000 (19:58 +0200)]
seg6: fix skb checksum evaluation in SRH encapsulation/insertion
Support for SRH encapsulation and insertion was introduced with
commit 6c8702c60b88 ("ipv6: sr: add support for SRH encapsulation and
injection with lwtunnels"), through the seg6_do_srh_encap() and
seg6_do_srh_inline() functions, respectively.
The former encapsulates the packet in an outer IPv6 header along with
the SRH, while the latter inserts the SRH between the IPv6 header and
the payload. Then, the headers are initialized/updated according to the
operating mode (i.e., encap/inline).
Finally, the skb checksum is calculated to reflect the changes applied
to the headers.
The IPv6 payload length ('payload_len') is not initialized
within seg6_do_srh_{inline,encap}() but is deferred in seg6_do_srh(), i.e.
the caller of seg6_do_srh_{inline,encap}().
However, this operation invalidates the skb checksum, since the
'payload_len' is updated only after the checksum is evaluated.
To solve this issue, the initialization of the IPv6 payload length is
moved from seg6_do_srh() directly into the seg6_do_srh_{inline,encap}()
functions and before the skb checksum update takes place.
Fixes: 6c8702c60b88 ("ipv6: sr: add support for SRH encapsulation and injection with lwtunnels") Reported-by: Paolo Abeni <pabeni@redhat.com> Link: https://lore.kernel.org/all/20220705190727.69d532417be7438b15404ee1@uniroma2.it Signed-off-by: Andrea Mayer <andrea.mayer@uniroma2.it> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Xiubo Li [Mon, 11 Jul 2022 04:11:21 +0000 (12:11 +0800)]
netfs: do not unlock and put the folio twice
check_write_begin() will unlock and put the folio when return
non-zero. So we should avoid unlocking and putting it twice in
netfs layer.
Change the way ->check_write_begin() works in the following two ways:
(1) Pass it a pointer to the folio pointer, allowing it to unlock and put
the folio prior to doing the stuff it wants to do, provided it clears
the folio pointer.
(2) Change the return values such that 0 with folio pointer set means
continue, 0 with folio pointer cleared means re-get and all error
codes indicating an error (no special treatment for -EAGAIN).
[ bagasdotme: use Sphinx code text syntax for *foliop pointer ]
Yang Yingliang [Thu, 7 Jul 2022 02:32:29 +0000 (10:32 +0800)]
xtensa: iss: fix handling error cases in iss_net_configure()
The 'pdev' and 'netdev' need to be released in error cases of
iss_net_configure().
Change the return type of iss_net_configure() to void, because it's
not used.
Fixes: 7282bee78798 ("[PATCH] xtensa: Architecture support for Tensilica Xtensa Part 8") Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
Max Filippov [Thu, 7 Jul 2022 07:23:16 +0000 (00:23 -0700)]
xtensa: iss/network: provide release() callback
Provide release() callback for the platform device embedded into struct
iss_net_private and registered in the iss_net_configure so that
platform_device_unregister could be called for it.
Max Filippov [Thu, 7 Jul 2022 05:57:35 +0000 (22:57 -0700)]
xtensa: iss/network: drop 'devices' list
There are two per-device lists in the ISS network driver: command line
parameters list and iss_net_private object list. The latter is only used
for duplicate checking in the function iss_net_setup where the former
should have been used.
Drop iss_net_private object list and associated code and use command
line parameters list in the iss_net_setup instead.
x86/entry: Remove UNTRAIN_RET from native_irq_return_ldt
UNTRAIN_RET is not needed in native_irq_return_ldt because RET
untraining has already been done at this point.
In addition, when the RETBleed mitigation is IBPB, UNTRAIN_RET clobbers
several registers (AX, CX, DX) so here it trashes user values which are
in these registers.
selftests/bpf: Return true/false (not 1/0) from bool functions
Return boolean values ("true" or "false") instead of 1 or 0 from bool
functions. This fixes the following warnings from coccicheck:
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:407:9-10: WARNING:
return of 0/1 in function 'decap_v4' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:389:9-10: WARNING:
return of 0/1 in function 'decap_v6' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:290:9-10: WARNING:
return of 0/1 in function 'encap_v6' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:264:9-10: WARNING:
return of 0/1 in function 'parse_tcp' with return type bool
tools/testing/selftests/bpf/progs/test_xdp_noinline.c:242:9-10: WARNING:
return of 0/1 in function 'parse_udp' with return type bool
BPF map name is limited to BPF_OBJ_NAME_LEN.
A map name is defined as being longer than BPF_OBJ_NAME_LEN,
it will be truncated to BPF_OBJ_NAME_LEN when a userspace program
calls libbpf to create the map. A pinned map also generates a path
in the /sys. If the previous program wanted to reuse the map，
it can not get bpf_map by name, because the name of the map is only
partially the same as the name which get from pinned path.
The syscall information below show that map name "process_pinned_map"
is truncated to "process_pinned_".
bpf(BPF_OBJ_GET, {pathname="/sys/fs/bpf/process_pinned_map",
bpf_fd=0, file_flags=0}, 144) = -1 ENOENT (No such file or directory)
This patch check that if the name of pinned map are the same as the
actual name for the first (BPF_OBJ_NAME_LEN - 1),
bpf map still uses the name which is included in bpf object.
Don Brace [Fri, 8 Jul 2022 18:48:06 +0000 (13:48 -0500)]
scsi: smartpqi: Update version to 2.1.18-045
Link: https://lore.kernel.org/r/165730608687.177165.11815510982277242966.stgit@brunhilda Reviewed-by: Gerry Morong <gerry.morong@microchip.com> Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Don Brace [Fri, 8 Jul 2022 18:48:01 +0000 (13:48 -0500)]
scsi: smartpqi: Update copyright to current year
Update copyright to current year.
Link: https://lore.kernel.org/r/165730608177.177165.13184715486635363193.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Allow user to override the default driver timeout for controller ready.
There are some rare configurations which require the driver to wait longer
than the normal 3 minutes for the controller to complete its bootup
sequence and be ready to accept commands from the driver.
The module parameter is:
ctrl_ready_timeout= { 0 | 30-1800 }
and specifies the timeout in seconds for the driver to wait for controller
ready. The valid range is 0 or 30-1800. The default value is 0, which
causes the driver to use a timeout of 180 seconds (3 minutes).
Link: https://lore.kernel.org/r/165730607666.177165.9221211345284471213.stgit@brunhilda Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kevin Barnett [Fri, 8 Jul 2022 18:47:51 +0000 (13:47 -0500)]
scsi: smartpqi: Update deleting a LUN via sysfs
Change removing a LUN using sysfs from an internal driver function
pqi_remove_all_scsi_devices() to using the .slave_destroy entry in the
scsi_host_template.
A LUN can be deleted via sysfs using this syntax:
echo 1 > /sys/block/sdX/device/delete
Link: https://lore.kernel.org/r/165730607154.177165.9723066932202995774.stgit@brunhilda Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mike McGowen [Fri, 8 Jul 2022 18:47:46 +0000 (13:47 -0500)]
scsi: smartpqi: Add module param to disable managed ints
Allow SMP affinity to be changeable by disabling managed interrupts.
On distributions where the driver is enabled for multi-queue support the
driver utilizes kernel managed interrupts, which automatically distributes
interrupts to all available CPUs and assigns SMP affinity.
On most distributions, the affinity can not be changed by the user.
This change will allow managed interrupts to be disabled by the user via a
module parameter while still allowing multi-queue support to function
properly.
Use the module parameter disable_managed_interrupts=1
Link: https://lore.kernel.org/r/165730606638.177165.12846020942931640329.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Kevin Barnett [Fri, 8 Jul 2022 18:47:41 +0000 (13:47 -0500)]
scsi: smartpqi: Fix RAID map race condition
Correct a rare stale RAID map access when performing AIO during a RAID
configuration change.
A race condition in the driver could cause it to access a stale RAID map
when a logical volume is reconfigured.
Modify the driver logic to invalidate a RAID map very early when a RAID
configuration change is detected and only switch to a new RAID map after
the driver detects that the RAID map has changed.
Link: https://lore.kernel.org/r/165730606128.177165.7671413443814750829.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: smartpqi: Fix DMA direction for RAID requests
Correct a SOP READ and WRITE DMA flags for some requests.
This update corrects DMA direction issues with SCSI commands removed from
the controller's internal lookup table.
Currently, SCSI READ BLOCK LIMITS (0x5) was removed from the controller
lookup table and exposed a DMA direction flag issue.
SCSI READ BLOCK LIMITS was recently removed from our controller lookup
table so the controller uses the respective IU flag field to set the DMA
data direction. Since the DMA direction is incorrect the FW never completes
the request causing a hang.
Some SCSI commands which use SCSI READ BLOCK LIMITS
* sg_map
* mt -f /dev/stX status
After updating controller firmware, users may notice their tape units
failing. This patch resolves the issue.
Also, the AIO path DMA direction is correct.
The DMA direction flag is a day-one bug with no reported BZ.
Fixes: 6c223761eb54 ("smartpqi: initial commit of Microsemi smartpqi driver") Link: https://lore.kernel.org/r/165730605618.177165.9054223644512926624.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Mahesh Rajashekhara <Mahesh.Rajashekhara@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Change method used to detect controller firmware crash during PQI reset.
PQI reset can fail with error -6 if firmware takes > 100ms to complete
reset.
Method used by driver to detect controller firmware crash during PQI was
incorrect in some cases.
Link: https://lore.kernel.org/r/165730605108.177165.1132931838384767071.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/165730604598.177165.9910276232981721083.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mike McGowen [Fri, 8 Jul 2022 18:47:20 +0000 (13:47 -0500)]
scsi: smartpqi: Add PCI ID for Adaptec SmartHBA 2100-8i
Add the PCI ID for (values in hex):
VID / DID / SVID / SDID
---- ---- ---- ----
Adaptec SmartHBA 2100-8i-o 9005 / 0285 / 9005 / 0659
Link: https://lore.kernel.org/r/165730604089.177165.17257514581321583667.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: smartpqi: Fix PCI control linkdown system hang
Fail all outstanding requests after a PCI linkdown.
Block access to device SCSI attributes during the following conditions:
"Cable pull" is called PQI_CTRL_SURPRISE_REMOVAL.
"PCIe Link Down" is called PQI_CTRL_GRACEFUL_REMOVAL.
Block access to device SCSI attributes during and in rare instances when
the controller goes offline.
Either outstanding requests or the access of SCSI attributes post linkdown
can lead to a hang.
Post linkdown, driver does not fail the outstanding requests leading to
long wait time before all the IOs eventually fail.
Also access of the SCSI attributes by host applications can lead to a
system hang.
Link: https://lore.kernel.org/r/165730603578.177165.4699352086827187263.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Sagar Biradar <sagar.biradar@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: smartpqi: Add driver support for multi-LUN devices
Add driver support for up to 256 LUNs per device.
Update AIO path to pass the appropriate LUN number for base-code to target
the correct LUN.
Update RAID IO path to pass the appropriate LUN number for FW to target the
correct LUN.
Pass the correct LUN number while doing a LUN reset.
Count the outstanding commands based on LUN number. While removing a
Multi-LUN device, wait for all outstanding commands to complete for all
LUNs.
Add Feature bit support.
Link: https://lore.kernel.org/r/165730603067.177165.14016422176841798336.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Kumar Meiyappan <Kumar.Meiyappan@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mike McGowen [Fri, 8 Jul 2022 18:47:05 +0000 (13:47 -0500)]
scsi: smartpqi: Close write read holes
Insert a minimum 1 millisecond delay after writing to a register before
reading from it.
SIS and PQI registers that can be both written to and read from can return
stale data if read from too soon after having been written to.
There is no read/write ordering or hazard detection on the inbound path to
the MSGU from the PCIe bus, therefore reads could pass writes.
Link: https://lore.kernel.org/r/165730602555.177165.11181012469428348394.stgit@brunhilda Reviewed-by: Scott Teel <scott.teel@microchip.com> Signed-off-by: Mike McGowen <mike.mcgowen@microchip.com> Co-developed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Link: https://lore.kernel.org/r/165730602045.177165.3720208650043407285.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Murthy Bhat <Murthy.Bhat@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: smartpqi: Add controller fw version to console log
Print controller firmware version to OS message log during driver
initialization or after OFA.
Link: https://lore.kernel.org/r/165730601536.177165.17698744242908911822.stgit@brunhilda Reviewed-by: Scott Benesh <scott.benesh@microchip.com> Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Mike McGowen <mike.mcgowen@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Gilbert Wu <Gilbert.Wu@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Mike McGowen [Fri, 8 Jul 2022 18:46:50 +0000 (13:46 -0500)]
scsi: smartpqi: Shorten drive visibility after removal
Check the response code returned from the LUN reset task management
function and if it indicates the LUN is not valid, do not retry.
Reduce rescan worker delay to 5 seconds for the event handler only.
The removal of a drive from the OS could have been delayed up to 30 seconds
after being physically pulled.
The driver was retrying a LUN reset 3 times even though the return code
indiciated the LUN was no longer valid. There was a 10 second delay between
each retry. Additionally, the rescan worker was scheduled to run 10 seconds
after the driver received the event.
Link: https://lore.kernel.org/r/165730601025.177165.9416869335174437006.stgit@brunhilda Reviewed-by: Scott Teel <scott.teel@microchip.com> Reviewed-by: Kevin Barnett <kevin.barnett@microchip.com> Signed-off-by: Mike McGowen <Mike.McGowen@microchip.com> Signed-off-by: Don Brace <don.brace@microchip.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Currently, the data flow of the max/min linkrate in the driver is
* in pm8001_get_lrate_mode():
hardcoded value ==> struct sas_phy
* in pm8001_bytes_dmaed():
struct pm8001_phy ==> struct sas_phy
* in pm8001_phy_control():
libsas data ==> struct pm8001_phy
Since pm8001_bytes_dmaed() follows pm8001_get_lrate_mode(), and the fields
in struct pm8001_phy are not initialized, sysfs
`/sys/class/sas_phy/phy-*/maximum_linkrate` always shows `Unknown`.
To fix the issue, change the dataflow to the following:
* in pm8001_phy_init():
initial value ==> struct pm8001_phy
* in pm8001_get_lrate_mode():
struct pm8001_phy ==> struct sas_phy
* in pm8001_phy_control():
libsas data ==> struct pm8001_phy
For negotiated linkrate, the current dataflow is:
* in pm8001_get_lrate_mode():
iomb data ==> struct asd_sas_phy ==> struct sas_phy
* in pm8001_bytes_dmaed():
struct asd_sas_phy ==> struct sas_phy
Since pm8001_bytes_dmaed() follows pm8001_get_lrate_mode(), the assignment
statements in pm8001_bytes_dmaed() are unnecessary and cleaned up.
Link: https://lore.kernel.org/r/20220707175210.528858-1-changyuanl@google.com Reviewed-by: Igor Pylypiv <ipylypiv@google.com> Acked-by: Jack Wang <jinpu.wang@ionos.com> Signed-off-by: Changyuan Lyu <changyuanl@google.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
In ufshcd_host_reset_and_restore(), ufshcd_set_clk_freq() is called to
scale clock rate. However, this did not call vops->clk_scale_notify() to
inform platform driver of clock change.
Call ufshcd_scale_clks() instead so that clock change can be properly
handled.
Link: https://lore.kernel.org/r/20220711144224.17916-2-powen.kao@mediatek.com Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Stanley Chu <stanley.chu@mediatek.com> Signed-off-by: Po-Wen Kao <powen.kao@mediatek.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: ufs: core: Drop loglevel of WriteBoost message
Commit '3b5f3c0d0548 ("scsi: ufs: core: Tidy up WB configuration code")'
changed the log level of the write boost enable/disable notification from
debug to info. This results in a lot of noise in the kernel log during
normal operation.
Drop it back to debug level to avoid this.
Link: https://lore.kernel.org/r/20220709000027.3929970-1-bjorn.andersson@linaro.org Fixes: 3b5f3c0d0548 ("scsi: ufs: core: Tidy up WB configuration code") Reviewed-by: Alim Akhtar <alim.akhtar@samsung.com> Acked-by: Bean Huo <beanhuo@micron.com> Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
scsi: fnic: Refactor code in fnic probe to initialize SCSI layer
Refactor code from fnic probe into a different function so that
scsi layer initialization code is grouped together.
Also, add log messages for better debugging.
Link: https://lore.kernel.org/r/20220707205155.692688-1-kartilak@cisco.com Co-developed-by: Gian Carlo Boffa <gcboffa@cisco.com> Signed-off-by: Gian Carlo Boffa <gcboffa@cisco.com> Co-developed-by: Arulprabhu Ponnusamy <arulponn@cisco.com> Signed-off-by: Arulprabhu Ponnusamy <arulponn@cisco.com> Signed-off-by: Karan Tilak Kumar <kartilak@cisco.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Michael Schmitz [Wed, 13 Jul 2022 07:49:13 +0000 (19:49 +1200)]
scsi: gvp11.c: Fix DMA mask calculation error
DMA masks given in the Zorro ID table don't contain the 2 byte alignment
quirk seen in the GVP11_XFER_MASK macro from gvp11.h so no need to account
for that.
DMA masks passed to dma_set_mask_and_coherent() must be 64 bit, add the
missing cast in the TO_DMA_MASK macro used to convert driver DMA masks to
DMA API masks.
Set the maximum time firmware should poll for a link.
If not set firmware could block CPU for a long time resulting
in mailbox failures. If link doesn't come up within 1second,
firmware will anyway notify the status as and when LINK comes up
Jakub Kicinski [Thu, 14 Jul 2022 03:16:03 +0000 (20:16 -0700)]
Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/net-queue
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2022-07-12
This series contains updates to ice driver only.
Paul fixes detection of E822 devices for firmware update and changes NVM
read for snapshot creation to be done in chunks as some systems cannot
read the entire NVM in the allotted time.
====================
Keoseong Park [Fri, 8 Jul 2022 05:20:06 +0000 (14:20 +0900)]
scsi: ufs: core: Remove UIC_HIBERN8_ENTER_RETRIES
Commit 4db7a2360597 ("scsi: ufs: Fix concurrency of error handler and other
error recovery paths") removed all callers of UIC_HIBERN8_ENTER_RETRIES.
Hence also remove the macro itself.
Ming Lei [Wed, 6 Jul 2022 12:59:42 +0000 (20:59 +0800)]
scsi: megaraid: Clear READ queue map's nr_queues
The megaraid SCSI driver sets set->nr_maps as 3 if poll_queues is > 0, and
blk-mq actually initializes each map's nr_queues as nr_hw_queues.
Consequently the driver has to clear READ queue map's nr_queues, otherwise
the queue map becomes broken if poll_queues is set as non-zero.
Link: https://lore.kernel.org/r/20220706125942.528533-1-ming.lei@redhat.com Fixes: 9e4bec5b2a23 ("scsi: megaraid_sas: mq_poll support") Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: sumit.saxena@broadcom.com Cc: chandrakanth.patil@broadcom.com Cc: linux-block@vger.kernel.org Cc: Hannes Reinecke <hare@suse.de> Reported-by: Guangwu Zhang <guazhang@redhat.com> Tested-by: Guangwu Zhang <guazhang@redhat.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Chanho Park [Wed, 6 Jul 2022 02:02:55 +0000 (11:02 +0900)]
scsi: ufs: ufs-exynos: Change ufs phy control sequence
Since commit 1599069a62c6 ("phy: core: Warn when phy_power_on is called
before phy_init"), the following warning has been reported:
phy_power_on was called before phy_init
To address this, we need to remove phy_power_on from exynos_ufs_phy_init()
and move it after phy_init. phy_power_off and phy_exit are also necessary
in exynos_ufs_remove().
Use after free is detected by kfence when disabling sriov. What was read
after being freed was vf->pci_dev: it was freed from pci_disable_sriov
and later read in efx_ef10_sriov_free_vf_vports, called from
efx_ef10_sriov_free_vf_vswitching.
Set the pointer to NULL at release time to not trying to read it later.
Reproducer and dmesg log (note that kfence doesn't detect it every time):
$ echo 1 > /sys/class/net/enp65s0f0np0/device/sriov_numvfs
$ echo 0 > /sys/class/net/enp65s0f0np0/device/sriov_numvfs
BUG: KFENCE: use-after-free read in efx_ef10_sriov_free_vf_vswitching+0x82/0x170 [sfc]
Use-after-free read at 0x00000000ff3c1ba5 (in kfence-#224):
efx_ef10_sriov_free_vf_vswitching+0x82/0x170 [sfc]
efx_ef10_pci_sriov_disable+0x38/0x70 [sfc]
efx_pci_sriov_configure+0x24/0x40 [sfc]
sriov_numvfs_store+0xfe/0x140
kernfs_fop_write_iter+0x11c/0x1b0
new_sync_write+0x11f/0x1b0
vfs_write+0x1eb/0x280
ksys_write+0x5f/0xe0
do_syscall_64+0x5c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
allocated by task 6771 on cpu 10 at 3137.860196s:
pci_alloc_dev+0x21/0x60
pci_iov_add_virtfn+0x2a2/0x320
sriov_enable+0x212/0x3e0
efx_ef10_sriov_configure+0x67/0x80 [sfc]
efx_pci_sriov_configure+0x24/0x40 [sfc]
sriov_numvfs_store+0xba/0x140
kernfs_fop_write_iter+0x11c/0x1b0
new_sync_write+0x11f/0x1b0
vfs_write+0x1eb/0x280
ksys_write+0x5f/0xe0
do_syscall_64+0x5c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
freed by task 6771 on cpu 12 at 3170.991309s:
device_release+0x34/0x90
kobject_cleanup+0x3a/0x130
pci_iov_remove_virtfn+0xd9/0x120
sriov_disable+0x30/0xe0
efx_ef10_pci_sriov_disable+0x57/0x70 [sfc]
efx_pci_sriov_configure+0x24/0x40 [sfc]
sriov_numvfs_store+0xfe/0x140
kernfs_fop_write_iter+0x11c/0x1b0
new_sync_write+0x11f/0x1b0
vfs_write+0x1eb/0x280
ksys_write+0x5f/0xe0
do_syscall_64+0x5c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
Fixes: 3c5eb87605e85 ("sfc: create vports for VFs and assign random MAC addresses") Reported-by: Yanghang Liu <yanghliu@redhat.com> Signed-off-by: Íñigo Huguet <ihuguet@redhat.com> Acked-by: Martin Habets <habetsm.xilinx@gmail.com> Link: https://lore.kernel.org/r/20220712062642.6915-1-ihuguet@redhat.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 02:05:07 +0000 (12:05 +1000)]
xfs: lockless buffer lookup
Now that we have a standalone fast path for buffer lookup, we can
easily convert it to use rcu lookups. When we continually hammer the
buffer cache with trylock lookups, we end up with a huge amount of
lock contention on the per-ag buffer hash locks:
This is basically hammering the pag->pag_buf_lock from lots of CPUs
doing trylocks at the same time. Most of the buffer trylock
operations ultimately fail after we've done the lookup, so we're
really hammering the buf hash lock whilst making no progress.
We can also see significant spinlock traffic on the same lock just
under normal operation when lots of tasks are accessing metadata
from the same AG, so let's avoid all this by converting the lookup
fast path to leverages the rhashtable's ability to do rcu protected
lookups.
We avoid races with the buffer release path by using
atomic_inc_not_zero() on the buffer hold count. Any buffer that is
in the LRU will have a non-zero count, thereby allowing the lockless
fast path to be taken in most cache hit situations. If the buffer
hold count is zero, then it is likely going through the release path
so in that case we fall back to the existing lookup miss slow path.
The slow path will then do an atomic lookup and insert under the
buffer hash lock and hence serialise correctly against buffer
release freeing the buffer.
The use of rcu protected lookups means that buffer handles now need
to be freed by RCU callbacks (same as inodes). We still free the
buffer pages before the RCU callback - we won't be trying to access
them at all on a buffer that has zero references - but we need the
buffer handle itself to be present for the entire rcu protected read
side to detect a zero hold count correctly.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 02:04:43 +0000 (12:04 +1000)]
xfs: remove a superflous hash lookup when inserting new buffers
Currently on the slow path insert we repeat the initial hash table
lookup before we attempt the insert, resulting in a two traversals
of the hash table to ensure the insert is valid. The rhashtable API
provides a method for an atomic lookup and insert operation, so we
can avoid one of the hash table traversals by using this method.
Adapted from a large patch containing this optimisation by Christoph
Hellwig.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 02:04:38 +0000 (12:04 +1000)]
xfs: reduce the number of atomic when locking a buffer after lookup
Avoid an extra atomic operation in the non-trylock case by only
doing a trylock if the XBF_TRYLOCK flag is set. This follows the
pattern in the IO path with NOWAIT semantics where the
"trylock-fail-lock" path showed 5-10% reduced throughput compared to
just using single lock call when not under NOWAIT conditions. So
make that same change here, too.
See commit 942491c9e6d6 ("xfs: fix AIM7 regression") for details.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
[hch: split from a larger patch] Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 02:04:31 +0000 (12:04 +1000)]
xfs: merge xfs_buf_find() and xfs_buf_get_map()
Now that we factored xfs_buf_find(), we can start separating into
distinct fast and slow paths from xfs_buf_get_map(). We start by
moving the lookup map and perag setup to _get_map(), and then move
all the specifics of the fast path lookup into xfs_buf_lookup()
and call it directly from _get_map(). We the move all the slow path
code to xfs_buf_find_insert(), which is now also called directly
from _get_map(). As such, xfs_buf_find() now goes away.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 02:02:46 +0000 (12:02 +1000)]
xfs: break up xfs_buf_find() into individual pieces
xfs_buf_find() is made up of three main parts: lookup, insert and
locking. The interactions with xfs_buf_get_map() require it to be
called twice - once for a pure lookup, and again on lookup failure
so the insert path can be run. We want to simplify this down a lot,
so split it into a fast path lookup, a slow path insert and a "lock
the found buffer" helper. This will then let us integrate these
operations more effectively into xfs_buf_get_map() in future
patches.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 01:47:42 +0000 (11:47 +1000)]
xfs: add in-memory iunlink log item
Now that we have a clean operation to update the di_next_unlinked
field of inode cluster buffers, we can easily defer this operation
to transaction commit time so we can order the inode cluster buffer
locking consistently.
To do this, we introduce a new in-memory log item to track the
unlinked list item modification that we are going to make. This
follows the same observations as the in-memory double linked list
used to track unlinked inodes in that the inodes on the list are
pinned in memory and cannot go away, and hence we can simply
reference them for the duration of the transaction without needing
to take active references or pin them or look them up.
This allows us to pass the xfs_inode to the transaction commit code
along with the modification to be made, and then order the logged
modifications via the ->iop_sort and ->iop_precommit operations
for the new log item type. As this is an in-memory log item, it
doesn't have formatting, CIL or AIL operational hooks - it exists
purely to run the inode unlink modifications and is then removed
from the transaction item list and freed once the precommit
operation has run.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de>
Dave Chinner [Thu, 14 Jul 2022 01:47:26 +0000 (11:47 +1000)]
xfs: add log item precommit operation
For inodes that are dirty, we have an attached cluster buffer that
we want to use to track the dirty inode through the AIL.
Unfortunately, locking the cluster buffer and adding it to the
transaction when the inode is first logged in a transaction leads to
buffer lock ordering inversions.
The specific problem is ordering against the AGI buffer. When
modifying unlinked lists, the buffer lock order is AGI -> inode
cluster buffer as the AGI buffer lock serialises all access to the
unlinked lists. Unfortunately, functionality like xfs_droplink()
logs the inode before calling xfs_iunlink(), as do various directory
manipulation functions. The inode can be logged way down in the
stack as far as the bmapi routines and hence, without a major
rewrite of lots of APIs there's no way we can avoid the inode being
logged by something until after the AGI has been logged.
As we are going to be using ordered buffers for inode AIL tracking,
there isn't a need to actually lock that buffer against modification
as all the modifications are captured by logging the inode item
itself. Hence we don't actually need to join the cluster buffer into
the transaction until just before it is committed. This means we do
not perturb any of the existing buffer lock orders in transactions,
and the inode cluster buffer is always locked last in a transaction
that doesn't otherwise touch inode cluster buffers.
We do this by introducing a precommit log item method. This commit
just introduces the mechanism; the inode item implementation is in
followup commits.
The precommit items need to be sorted into consistent order as we
may be locking multiple items here. Hence if we have two dirty
inodes in cluster buffers A and B, and some other transaction has
two separate dirty inodes in the same cluster buffers, locking them
in different orders opens us up to ABBA deadlocks. Hence we sort the
items on the transaction based on the presence of a sort log item
method.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de>
Dave Chinner [Thu, 14 Jul 2022 01:46:59 +0000 (11:46 +1000)]
xfs: combine iunlink inode update functions
Combine the logging of the inode unlink list update into the
calling function that looks up the buffer we end up logging. These
do not need to be separate functions as they are both short, simple
operations and there's only a single call path through them. This
new function will end up being the core of the iunlink log item
processing...
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 01:46:43 +0000 (11:46 +1000)]
xfs: double link the unlinked inode list
Now we have forwards traversal via the incore inode in place, we now
need to add back pointers to the incore inode to entirely replace
the back reference cache. We use the same lookup semantics and
constraints as for the forwards pointer lookups during unlinks, and
so we can look up any inode in the unlinked list directly and update
the list pointers, forwards or backwards, at any time.
The only wrinkle in converting the unlinked list manipulations to
use in-core previous pointers is that log recovery doesn't have the
incore inode state built up so it can't just read in an inode and
release it to finish off the unlink. Hence we need to modify the
traversal in recovery to read one inode ahead before we
release the inode at the head of the list. This populates the
next->prev relationship sufficient to be able to replay the unlinked
list and hence greatly simplify the runtime code.
This recovery algorithm also requires that we actually remove inodes
from the unlinked list one at a time as background inode
inactivation will result in unlinked list removal racing with the
building of the in-memory unlinked list state. We could serialise
this by holding the AGI buffer lock when constructing the in memory
state, but all that does is lockstep background processing with list
building. It is much simpler to flush the inodegc immediately after
releasing the inode so that it is unlinked immediately and there is
no races present at all.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de>
Dave Chinner [Thu, 14 Jul 2022 01:43:09 +0000 (11:43 +1000)]
xfs: introduce xfs_iunlink_lookup
When an inode is on an unlinked list during normal operation, it is
guaranteed to be pinned in memory as it is either referenced by the
current unlink operation or it has a open file descriptor that
references it and has it pinned in memory. Hence to look up an inode
on the unlinked list, we can do a direct inode cache lookup and
always expect the lookup to succeed.
Add a function to do this lookup based on the agino that we use to
link the chain of unlinked inodes together so we can begin the
conversion the unlinked list manipulations to use in-memory inodes
rather than inode cluster buffers and remove the backref cache.
Use this lookup function to replace the on-disk inode buffer walk
when removing inodes from the unlinked list with an in-core inode
unlinked list walk.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 01:42:39 +0000 (11:42 +1000)]
xfs: refactor xlog_recover_process_iunlinks()
For upcoming changes to the way inode unlinked list processing is
done, the structure of recovery needs to change slightly. We also
really need to untangle the messy error handling in list recovery
so that actions like emptying the bucket on inode lookup failure
are associated with the bucket list walk failing, not failing
to look up the inode.
Refactor the recovery code now to keep the re-organisation seperate
to the algorithm changes.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Dave Chinner [Thu, 14 Jul 2022 01:38:54 +0000 (11:38 +1000)]
xfs: track the iunlink list pointer in the xfs_inode
Having direct access to the i_next_unlinked pointer in unlinked
inodes greatly simplifies the processing of inodes on the unlinked
list. We no longer need to look up the inode buffer just to find
next inode in the list if the xfs_inode is in memory. These
improvements will be realised over upcoming patches as other
dependencies on the inode buffer for unlinked list processing are
removed.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Reviewed-by: Christoph Hellwig <hch@lst.de>
Dave Chinner [Thu, 14 Jul 2022 01:36:40 +0000 (11:36 +1000)]
xfs: factor the xfs_iunlink functions
Prep work that separates the locking that protects the unlinked list
from the actual operations being performed. This also helps document
the fact they are performing list insert and remove operations. No
functional code change.
Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Zhang Yi [Thu, 14 Jul 2022 01:36:36 +0000 (11:36 +1000)]
xfs: flush inode gc workqueue before clearing agi bucket
In the procedure of recover AGI unlinked lists, if something bad
happenes on one of the unlinked inode in the bucket list, we would call
xlog_recover_clear_agi_bucket() to clear the whole unlinked bucket list,
not the unlinked inodes after the bad one. If we have already added some
inodes to the gc workqueue before the bad inode in the list, we could
get below error when freeing those inodes, and finaly fail to complete
the log recover procedure.
XFS (ram0): Internal error xfs_iunlink_remove at line 2456 of file
fs/xfs/xfs_inode.c. Caller xfs_ifree+0xb0/0x360 [xfs]
The problem is xlog_recover_clear_agi_bucket() clear the bucket list, so
the gc worker fail to check the agino in xfs_verify_agino(). Fix this by
flush workqueue before clearing the bucket.
Fixes: ab23a7768739 ("xfs: per-cpu deferred inode inactivation queues") Signed-off-by: Zhang Yi <yi.zhang@huawei.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Darrick J. Wong <djwong@kernel.org> Signed-off-by: Dave Chinner <david@fromorbit.com>
KVM: VMX: Update PT MSR intercepts during filter change iff PT in host+guest
Update the Processor Trace (PT) MSR intercepts during a filter change if
and only if PT may be exposed to the guest, i.e. only if KVM is operating
in the so called "host+guest" mode where PT can be used simultaneously by
both the host and guest. If PT is in system mode, the host is the sole
owner of PT and the MSRs should never be passed through to the guest.
Luckily the missed check only results in unnecessary work, as select RTIT
MSRs are passed through only when RTIT tracing is enabled "in" the guest,
and tracing can't be enabled in the guest when KVM is in system mode
(writes to guest.MSR_IA32_RTIT_CTL are disallowed).
KVM: selftests: Use the common cpuid() helper in cpu_vendor_string_is()
Use cpuid() to get CPUID.0x0 in cpu_vendor_string_is(), thus eliminating
the last open coded usage of CPUID (ignoring debug_regs.c, which emits
CPUID from the guest to trigger a VM-Exit and doesn't actually care about
the results of CPUID).
KVM: selftests: Clean up requirements for XFD-aware XSAVE features
Provide informative error messages for the various checks related to
requesting access to XSAVE features that are buried behind XSAVE Feature
Disabling (XFD).
Opportunistically rename the helper to have "require" in the name so that
it's somewhat obvious that the helper may skip the test.
KVM: selftests: Skip AMX test if ARCH_REQ_XCOMP_GUEST_PERM isn't supported
Skip the AMX test instead of silently returning if the host kernel
doesn't support ARCH_REQ_XCOMP_GUEST_PERM. KVM didn't support XFD until
v5.17, so it's extremely unlikely allowing the test to run on a pre-v5.15
kernel is the right thing to do.
KVM: selftests: Check KVM's supported CPUID, not host CPUID, for XFD
Use kvm_cpu_has() to check for XFD supported in vm_xsave_req_perm(),
simply checking host CPUID doesn't guarantee KVM supports AMX/XFD.
Opportunistically hoist the check above the bit check; if XFD isn't
supported, it's far better to get a "not supported at all" message, as
opposed to a "feature X isn't supported" message".
KVM: selftests: Rename kvm_get_supported_cpuid_index() to __..._entry()
Rename kvm_get_supported_cpuid_index() to __kvm_get_supported_cpuid_entry()
to better show its relationship to kvm_get_supported_cpuid_entry(), and
because the helper returns a CPUID entry, not the index of an entry.
KVM: selftests: Drop unnecessary use of kvm_get_supported_cpuid_index()
Use kvm_get_supported_cpuid_entry() instead of
kvm_get_supported_cpuid_index() when passing in '0' for the index, which
just so happens to be the case in all remaining users of
kvm_get_supported_cpuid_index() except kvm_get_supported_cpuid_entry().
Keep the helper as there may be users in the future, and it's not doing
any harm.
KVM: selftests: Add this_cpu_has() to query X86_FEATURE_* via cpuid()
Add this_cpu_has() to query an X86_FEATURE_* via cpuid(), i.e. to query a
feature from L1 (or L2) guest code. Arbitrarily select the AMX test to
be the first user.
KVM: selftests: Set input function/index in raw CPUID helper(s)
Set the function/index for CPUID in the helper instead of relying on the
caller to do so. In addition to reducing the risk of consuming an
uninitialized ECX, having the function/index embedded in the call makes
it easier to understand what is being checked.
KVM: selftests: Make get_supported_cpuid() returns "const"
Tag the returned CPUID pointers from kvm_get_supported_cpuid(),
kvm_get_supported_hv_cpuid(), and vcpu_get_supported_hv_cpuid() "const"
to prevent reintroducing the broken pattern of modifying the static
"cpuid" variable used by kvm_get_supported_cpuid() to cache the results
of KVM_GET_SUPPORTED_CPUID.
KVM: selftests: Use vcpu_{set,clear}_cpuid_feature() in nVMX state test
Use vcpu_{set,clear}_cpuid_feature() to toggle nested VMX support in the
vCPU CPUID module in the nVMX state test. Drop CPUID_VMX as there are
no longer any users.
KVM: selftests: Use vCPU's CPUID directly in Hyper-V test
Use the vCPU's persistent CPUID array directly when manipulating the set
of exposed Hyper-V CPUID features. Drop set_cpuid() to route all future
modification through the vCPU helpers; the Hyper-V features test was the
last user.
KVM: selftests: Use vcpu_get_cpuid_entry() in PV features test (sort of)
Add a new helper, vcpu_clear_cpuid_entry(), to do a RMW operation on the
vCPU's CPUID model to clear a given CPUID entry, and use it to clear
KVM's paravirt feature instead of operating on kvm_get_supported_cpuid()'s
static "cpuid" variable. This also eliminates a user of
the soon-be-defunct set_cpuid() helper.
KVM: selftests: Add and use helper to set vCPU's CPUID maxphyaddr
Add a helper to set a vCPU's guest.MAXPHYADDR, and use it in the test
that verifies the emulator returns an error on an unknown instruction
when KVM emulates in response to an EPT violation with a GPA that is
legal in hardware but illegal with respect to the guest's MAXPHYADDR.
Add a helper even though there's only a single user at this time. Before
its removal, mmu_role_test also stuffed guest.MAXPHYADDR, and the helper
provides a small amount of clarity.
More importantly, this eliminates a set_cpuid() user and an instance of
modifying kvm_get_supported_cpuid()'s static "cpuid".
KVM: selftests: Use vm->pa_bits to generate reserved PA bits
Use vm->pa_bits to generate the mask of physical address bits that are
reserved in page table entries. vm->pa_bits is set when the VM is
created, i.e. it's guaranteed to be valid when populating page tables.
KVM: selftests: Add helpers to get and modify a vCPU's CPUID entries
Add helpers to get a specific CPUID entry for a given vCPU, and to toggle
a specific CPUID-based feature for a vCPU. The helpers will reduce the
amount of boilerplate code needed to tweak a vCPU's CPUID model, improve
code clarity, and most importantly move tests away from modifying the
static "cpuid" returned by kvm_get_supported_cpuid().
KVM: selftests: Rename and tweak get_cpuid() to get_cpuid_entry()
Rename get_cpuid() to get_cpuid_entry() to better reflect its behavior.
Leave set_cpuid() as is to avoid unnecessary churn, that helper will soon
be removed entirely.
Oppurtunistically tweak the implementation to avoid using a temporary
variable in anticipation of taggin the input @cpuid with "const".
KVM: selftests: Don't use a static local in vcpu_get_supported_hv_cpuid()
Don't use a static variable for the Hyper-V supported CPUID array, the
helper unconditionally reallocates the array on every invocation (and all
callers free the array immediately after use). The array is intentionally
recreated and refilled because the set of supported CPUID features is
dependent on vCPU state, e.g. whether or not eVMCS has been enabled.
Cache a vCPU's CPUID information in "struct kvm_vcpu" to allow fixing the
mess where tests, often unknowingly, modify the global/static "cpuid"
allocated by kvm_get_supported_cpuid().
Add vcpu_init_cpuid() to handle stuffing an entirely different CPUID
model, e.g. during vCPU creation or when switching to the Hyper-V enabled
CPUID model. Automatically refresh the cache on vcpu_set_cpuid() so that
any adjustments made by KVM are always reflected in the cache. Drop
vcpu_get_cpuid() entirely to force tests to use the cache, and to allow
adding e.g. vcpu_get_cpuid_entry() in the future without creating a
conflicting set of APIs where vcpu_get_cpuid() does KVM_GET_CPUID2, but
vcpu_get_cpuid_entry() does not.
Opportunistically convert the VMX nested state test and KVM PV test to
manipulating the vCPU's CPUID (because it's easy), but use
vcpu_init_cpuid() for the Hyper-V features test and "emulator error" test
to effectively retain their current behavior as they're less trivial to
convert.
KVM: selftests: Split out kvm_cpuid2_size() from allocate_kvm_cpuid2()
Split out the computation of the effective size of a kvm_cpuid2 struct
from allocate_kvm_cpuid2(), and modify both to take an arbitrary number
of entries. Future commits will add caching of a vCPU's CPUID model, and
will (a) be able to precisely size the entries array, and (b) will need
to know the effective size of the struct in order to copy to/from the
cache.
Expose the helpers so that the Hyper-V Features test can use them in the
(somewhat distant) future. The Hyper-V test very, very subtly relies on
propagating CPUID info across vCPU instances, and will need to make a
copy of the previous vCPU's CPUID information when it switches to using
the per-vCPU cache. Alternatively, KVM could provide helpers to
duplicate and/or copy a kvm_cpuid2 instance, but each is literally a
single line of code if the helpers are exposed, and it's not like the
size of kvm_cpuid2 is secret knowledge.
KVM: selftests: Verify that kvm_cpuid2.entries layout is unchanged by KVM
In the CPUID test, verify that KVM doesn't modify the kvm_cpuid2.entries
layout, i.e. that the order of entries and their flags is identical
between what the test provides via KVM_SET_CPUID2 and what KVM returns
via KVM_GET_CPUID2.
Asserting that the layouts match simplifies the test as there's no need
to iterate over both arrays.
KVM: selftests: Remove the obsolete/dead MMU role test
Remove the MMU role test, which was made obsolete by KVM commit feb627e8d6f6 ("KVM: x86: Forbid KVM_SET_CPUID{,2} after KVM_RUN"). The
ongoing costs of keeping the test updated far outweigh any benefits,
e.g. the test _might_ be useful as an example or for documentation
purposes, but otherwise the test is dead weight.
Use kvm_cpu_has() in the AMX test instead of open coding equivalent
functionality using kvm_get_supported_cpuid_entry() and
kvm_get_supported_cpuid_index().
KVM: selftests: Check for _both_ XTILE data and cfg in AMX test
Check for _both_ XTILE data and cfg support in the AMX test instead of
checking for _either_ feature. Practically speaking, no sane CPU or vCPU
will support one but not the other, but the effective "or" behavior is
subtle and technically incorrect.
KVM: selftests: Drop redundant vcpu_set_cpuid() from PMU selftest
Drop a redundant vcpu_set_cpuid() from the PMU test. The vCPU's CPUID is
set to KVM's supported CPUID by vm_create_with_one_vcpu(), which was also
true back when the helper was named vm_create_default().
projects / people / ms / linux.git / log
