In Linux v6.16, commit 41cb08555c41 ("treewide, timers: Rename
from_timer() to timer_container_of()") renamed 'from_timer()' to
'timer_container_of()'. Given that this is a macro we can simplfy see if
the macro 'timer_container_of' is defined and if so use this otherwise
fall back to 'from_timer()'. Update the necessary drivers to fix the
build for Linux v6.16.
JIRA LINQPJ14-60
Change-Id: I7f622b5d046a92da2ade755e6a697c1810f61275
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3383387
(cherry picked from commit 320ec84efd492c0c2711c69104fabc30b4f15ecb)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3461850
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
- The Linux BSP secutiry requirement "LINUXBSP70-REQ-223"
mandates the memory to be initialized before use. So, updated
the driver to use zero initialized memory allocation API's.
- Removed the IVC reset done during resume.
Jira SSV-12846
Change-Id: Iac9bd11edb520584206797446b8f3a59d20b2453
Signed-off-by: Sanjith T D <std@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3354502
Reviewed-by: Sreenivas Velpula <svelpula@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Vipin Kumar <vipink@nvidia.com>
In Linux v6.15, the timer APIs hrtimer_init() and del_timer() have been
removed. The hrtimer_setup() was added in Linux v6.13 to replace
hrtimer_init() and hrtimer_init() have finally been removed. The
functions del_timer()/del_timer_sync() were renamed to
timer_delete()/timer_delete_sync() in Linux v6.15. Use conftest to
detect these changes and update the drivers as necessary.
JIRA LINQPJ14-47
Change-Id: Id3994900384aad4b91155507cda91e04898ab12c
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3336168
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
In order to provide separate access control for
Device Management and Firmware Update IOCTL's,
separate device nodes are created.
- /dev/vblkdevX.ctl is for Device Management related IOCTL's
- /dev/vblkdevX.ffu is for Firmware Update related IOCTL's
Removed the old approach of checking for access permission
inside the driver.
Jira SSV-12846
Change-Id: Id34d37a34eb1b988559d5dc3aec0b2291dcecc8f
Signed-off-by: Sanjith T D <std@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3324229
Reviewed-by: Vipin Kumar <vipink@nvidia.com>
Reviewed-by: Sreenivas Velpula <svelpula@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Previous code was using the struct request.completion_data
to pass the struct ioctl_request to the processing thread.
This member variable is removed from kernel v6.5.
Change contains:
[1] Moving this variable from struct request to struct struct vblk_dev.
[2] Copyright Banner is updated
This makes the passing of the struct ioctl_request independent of
linux kernel transition
Bug 4311184
Change-Id: I3dd2c4ee9861516db770adcba345362395bf3fa3
Signed-off-by: gokull <gokull@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3314801
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Method to configure the struct queue_limits has changed
from kernel v6.11.
The following API's are removed from v6.11
[1] blk_queue_max_secure_erase_sectors
[2] blk_queue_max_discard_sectors
[3] blk_queue_write_cache
[4] blk_queue_logical_block_size
[5] blk_queue_physical_block_size
[6] blk_queue_max_hw_sectors
[7] The semantics of the flag QUEUE_FLAG_NONROT
is set as default value for queues, so this flag
is removed in v6.11.
[8] The flag BLK_MQ_F_SHOULD_MERGE is not valid in
v6.14
The above API's are replaced by direct
assignment to the queue_limits and use the new API
queue_limits_set. This API takes the incoming
values, validates, populates the other members
of the struct accordingly and assigns the
provided queue_limits to the corresponding queue.
The change contains:
[1] Addition of new approach that is compatible
with v6.11
[2] Addition of a macro (NV_BLOCK_USE_QUEUE_LIMITS_SET)
for backward compatibility
[3] This is a special case where both the variable
and the function needs to be present for the macro to be
enabled. The changes were made accordingly.
[4] The flag BLK_MQ_F_SHOULD_MERGE is protected by
macro
Bug 4311184
Signed-off-by: gokull <gokull@nvidia.com>
Change-Id: Ida2c12de0977f6405e7cc7dec946cf0f64413c97
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3303900
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Currently vblk driver uses wait_for_completion() for the requests from block layer or responses from Storage Server. wait_for_completion() make the current waiter to be un-interrutable. But linux top utility loadavg metric includes the processes both in running state and uninterruptable sate.
Replace wait_for_completion() with wait_for_completion_interruptible() in driver, so that vblk wait will not be included in loadavg.
This changes is made as per customer request. It is for only cosmetic or for accounting purposes and no effect in functionality.
Bug 4767390
Change-Id: I2b885a1623b7d2d3a7af00951c5421dacf9d3d26
Signed-off-by: Vikram Kanigiri <vkanigiri@nvidia.com>
(cherry picked from commit ff34c49236bf57542eae09a7e7f7189ac8bddd13)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3219307
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Sreenivas Velpula <svelpula@nvidia.com>
Reviewed-by: Vipin Kumar <vipink@nvidia.com>
(cherry picked from commit 2784d7b651cea6eb22195443d2cf43593084324b)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3262249
Reviewed-by: Sumeet Gupta <sumeetg@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: Manish Bhardwaj <mbhardwaj@nvidia.com>
Tested-by: Manish Bhardwaj <mbhardwaj@nvidia.com>
Reviewed-by: Gokul Vasan L J <gokull@nvidia.com>
The Tegra virtual storage driver fails to build for Linux v6.11 because
of the following issues:
1. The functions blk_queue_logical_block_size(),
blk_queue_physical_block_size() and blk_queue_max_discard_sectors()
have been removed.
2. The function blk_queue_write_cache() has too many arguments for
Linux v6.11.
3. The flag QUEUE_FLAG_NONROT is not defined for Linux v6.11.
Add a dummy driver for this driver for Linux v6.11 until this driver is
fixed.
Bug 4311184
Bug 4749580
Change-Id: I56f811d6ee29dd0025a5dc199ae84e3fc0f37b21
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3162886
Reviewed-by: Laxman Dewangan <ldewangan@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
The function blk_queue_max_hw_sectors() was removed in Linux v6.10 and
so add a test to conftest to see if this function is present or not to
allow the driver to build. The virtual-storage driver is currently
broken for Linux v5.19+ kernel due to other incompatibilities with these
newer kernels, but has been updated so that it still builds. Therefore,
allow the driver to build for Linux v6.10 while the driver is fixed
properly for these newer kernels.
Bug 4311184
Bug 4593750
Change-Id: Ic4054c87acdf3c77f03ff94fbdde513d84fe229b
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3142216
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Upstream commit 690da22dbfa8 ("asm-generic/io.h: kill vmalloc.h
dependency") removed the vmalloc.h header file from io.h and this breaks
building various drivers with the latest -next kernels. Fix this
by ensuring vmalloc.h is included and in most cases slab is not actually
needed and so remove this where possible. Note that it is fine to make
this change for all current supported kernels.
Bug 4593750
Change-Id: I003d1302bda226d356467e6ede99949b2716940a
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3141984
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
In Linux v6.9-rc1, the function blk_mq_init_queue() was renamed
blk_mq_alloc_queue() and the arguments passed to the function were
updated. Add a test to conftest to detect if the function
blk_mq_alloc_queue() is present and update the virtual-storage driver
accordingly.
Bug 4471899
Change-Id: I7dc937eaad27445b1c140c57aafd36f4a4b769ba
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3095926
Reviewed-by: Paritosh Dixit <paritoshd@nvidia.com>
In Linux v5.17, the GENHD_FL_EXT_DEVT and GENHD_FL_NO_PART_SCAN were
removed and in Linux v5.19, the flags QUEUE_FLAG_SECERASE and
QUEUE_FLAG_DISCARD were removed. Rather than using kernel version
checks, simply check if the flag is defined. This is beneficial for
working with 3rd party Linux kernels that may have backported changes
from upstream.
Although the Tegra virtual-storage can be compiled for v5.19 kernels, it
still needs to be fixed properly for v5.19 kernels. Add some dev_WARN
prints to warn if these flags are not supported so that it will be clear
that this is not expected to work.
Bug 4119327
Bug 4311184
Change-Id: I7f7585b238ad45b26fb1d0df42b338c904ce87e7
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2992552
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Rather than using kernel version checks to determine which kernel APIs
to use, add the necessary tests to the conftest script to determine
which kernel APIs are present in the kernel.
This is beneficial for working with 3rd party Linux kernels that may
have back-ported upstream changes into their kernel and so the kernel
version checks do not work.
Bug 4221847
Change-Id: Iec2c793ce408dab1cf7e5118c019dfe656dfa87c
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2992630
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
The function blk_mq_alloc_request() has been exported in the Linux
kernel since v3.16 and the function blk_mq_free_request() has been
exported in the Linux kernel since v3.19. The function
Since Linux v5.0, blk_get_request() has always called
blk_mq_alloc_request() and blk_put_request() has always called
blk_mq_free_request(). Given that kernels newer than v5.0 are now
supported drop the version check around blk_mq_alloc_request() and
blk_mq_free_request() and always use these functions.
Bug 4119327
Change-Id: I018687b340acaa9617c81551efb436a0439b14f1
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2991457
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Building the Tegra virtual storage driver for v5.14 kernels fails with
the following error ...
drivers/block/tegra_virt_storage/tegra_hv_vblk.c:1236:48: error:
passing argument 1 of '__alloc_disk_node' makes integer from pointer
without a cast [-Werror=int-conversion]
vblkdev->gd = __alloc_disk_node(vblkdev->queue, NUMA_NO_NODE);
~~~~~~~^~~~~~~
|
struct request_queue *
Fix this by correcting the arguments passed to __alloc_disk_node() for
pre-v5.15 kernels.
Bug 3820317
Change-Id: I2f208acb3ef0a009d877ab46fb110a87968862fe
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2978392
Reviewed-by: Laxman Dewangan <ldewangan@nvidia.com>
Below race condition can happen during boot
1) Server calls sivc_reset()
2) Storage client driver has IVC event handler registered, so it gets the event
and it calls ivc_notified()
3) Server gets ivc event, it calls ivc_notified()
( this makes server local state to established )
4) Storage cliend driver also goes to established state
5) After this, storage client driver continued with probe() and
calls ivc_reset() which results in disconnecting the
already established connection.
To fix this race condition, perform ivc_reset in storage client
driver right after tegra_hv_ivc_reserve() before registering
the IVC IRQ handler.
Bug 4125055
Change-Id: I36e6db3479f34fb649bdb8d890f24b1783d10cb9
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2943766
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Manish Bhardwaj <mbhardwaj@nvidia.com>
Reviewed-by: Sanjith T D <std@nvidia.com>
Reviewed-by: Vipin Kumar <vipink@nvidia.com>
Tested-by: Manish Bhardwaj <mbhardwaj@nvidia.com>
1. If vs_blk_dev_config.req_ops_supported has only DISCARD
support enabled, then REQ_OP_DISCARD should be mapped to
DISCARD for UFS.
2. If vs_blk_dev_config.req_ops_supported has only SECURE_ERASE
support enabled, then REQ_OP_DISCARD and REQ_OP_SECURE_ERASE
both should be mapped to SECURE_ERASE for UFS.
3. If vs_blk_dev_config.req_ops_supported has only ERASE
support enabled, then REQ_OP_DISCARD and REQ_OP_SECURE_ERASE
both should be mapped to ERASE for UFS.
Bug 4176555
Signed-off-by: Manish Bhardwaj <mbhardwaj@nvidia.com>
Change-Id: I01599df9ab93525120106dfabf2d345ab8b64770
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/2920433
Reviewed-by: Sanjith T D <std@nvidia.com>
Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com>
Reviewed-by: svc-mobile-cert <svc-mobile-cert@nvidia.com>
Reviewed-by: Sandeep Trasi <strasi@nvidia.com>
GVS: Gerrit_Virtual_Submit <buildbot_gerritrpt@nvidia.com>
