Commit graph

1,031,458 commits

Author SHA1 Message Date
Sathishkumar S
3fa8f89d72 drm/amdgpu: enable smart shift on dGPU (v5)
enable smart shift on dGPU if it is part of HG system and
the platform supports ATCS method to handle power shift.

V2: avoid psc updates in baco enter and exit (Lijo)
    fix alignment (Shashank)
V3: rebased on unified ATCS handling. (Alex)
V4: check for return value and warn on failed update (Shashank)
    return 0 if device does not support smart shift.  (Lizo)
V5: rebased on ATPX/ATCS structures global (Alex)

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Shashank Sharma <shashank.sharma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
19a1d9350b drm/amdgpu: flush gart changes after all BO recovery
Don't flush gart changes after recovering each BO instead
do it after recovering all the BOs. Flishing gart also needed
for amdgpu_ttm_alloc_gart().

v4: use containerof to retrieve adev struct.
v3: rename amdgpu_gart_tlb_flush() -> amdgpu_gart_invalidate_tlb().
v2: abstract out gart tlb flushing logic to amdgpu_gart.c

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
c7b9aa7a92 drm/amdgpu: do not allocate entries separately
Allocate PD/PT entries while allocating VM BOs and use that
instead of allocating those entries separately.

v2: create a new var for num entries.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
9c3fec688f drm/amdgpu: remove unused code
Remove unused code related to shadow BO.

v2: removing shadow bo ptr from base class.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
59276f056f drm/amdgpu: switch to amdgpu_bo_vm for vm code
The subclass, amdgpu_bo_vm is intended for PT/PD BOs which are also
shadowed, so switch to amdgpu_bo_vm BO for PT/PD BOs.

v4: update amdgpu_vm_update_funcs to accept amdgpu_bo_vm.
v3: simplify code.
    check also if shadow bo exist instead of checking bo only type.
v2: squash three related patches.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
1fdc79f6f9 drm/admgpu: add two shadow BO helper functions
Add amdgpu_bo_add_to_shadow_list() to handle shadow list
additions and amdgpu_bo_shadowed() to check if a BO is shadowed.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
2a675640bc drm/amdgpu: move shadow bo validation to VM code
Do the shadow bo validation in the VM code as
VM code knows/owns shadow BOs.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
6fdd6f4aa5 drm/amdgpu: add amdgpu_bo_vm bo type
Add new BO subclass that will be used by amdgpu vm code.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:38 -04:00
Nirmoy Das
ae4c0d7674 drm/amdgpu: make sure we unpin the UVD BO
Releasing pinned BOs is illegal now. UVD 6 was missing from:
commit 2f40801dc5 ("drm/amdgpu: make sure we unpin the UVD BO")

Fixes: 2f40801dc5 ("drm/amdgpu: make sure we unpin the UVD BO")
Cc: stable@vger.kernel.org
Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:55:14 -04:00
Xiaojian Du
e0172928ae drm/amd/pm: retain the fine grain tuning parameters after resume
This patch is to retain the fine grain tuning parameters after resume for
legacy APU, it will cover Raven/Raven2/Picasso.

Signed-off-by: Xiaojian Du <Xiaojian.Du@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:46:05 -04:00
Nirmoy Das
bcd743741f drm/amd/display: WARN_ON cleanups
Use WARN_ON() inside if-condition when possible.

v2: Squash in coding style fix (Alex)

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:46:04 -04:00
Dave Airlie
43ed3c6c78 drm-misc-next for 5.14:
UAPI Changes:
 
  * Use DRM driver names for fbdev
 
 Cross-subsystem Changes:
 
 Core Changes:
 
  * Fix leaked DMA handles
 
  * Improve documentation around DRM_CLIENT_CAP_*
 
  * Cleanups
 
  * dp_mst: Use kHz as link-rate unit during init
 
  * fourcc: Remove drm_gem_format_name() and drm_format_name_buf
 
  * gem-cma: Fix mmap for buffers with write combining
 
  * ttm: Don't override pre-set vm_ops; ttm_bo_mmap() removal and cleanups
 
 Driver Changes:
 
  * drm/amdgpu: Fix hot unplug during suspend; Implement mmap as GEM object
    function; Use %p4cc format-string modifier; Cleanups
 
  * drm/bridge: Cdns: Fix PM reference leak, Cleanups; Lt8912b: Fix Coccinelle
    warnings; Fix Kconfig dependencies; Fixes and cleanups
 
  * drm/hisilicon/kirin: Cleanups
 
  * drm/nouveau: Implement mmap as GEM object function
 
  * drm/radeon: Implement mmap as GEM object function
 
  * drm/rockchip: Remove generic drivers during init; Add scaling for RK3036
    win1; Fix missing registers for RK3066 and 3188; Add alpha support for
    RK3036, RK3066, RK3126 and RK3188; Fixes and cleanups
 
  * drm/simpledrm: Use %p4cc: format-string modifier
 
  * drm/vmwgfx: Cleanups
 
  * fbdev/matrox: Use modern module_init()
 -----BEGIN PGP SIGNATURE-----
 
 iQEzBAABCAAdFiEEchf7rIzpz2NEoWjlaA3BHVMLeiMFAmC2S24ACgkQaA3BHVML
 eiPiSwgAuWKdl4QNoMquYwORm96NHVQaWxtrLzc7QanEKz4eyMJ7MQlC1ifX0101
 5qEkCx4it3wZPmMyN1GZtqwvu38zyXqPTwqA5f6Llb7xNiYxc2e69xT/K/zm3Qs0
 sSC7pUeQKtVfye9eciOF9n/TMfnFqsePCUUg7QUqv/V/tb43dm5OVo6bDVzGKlNY
 wmlmt3vfdxMWnFVrB3qh/3Um3KAs+lLiMnMX33hqbDkGXeI0rBAHC4MAkcOdWZzl
 CFIjKfrc5aQUJXFBA2t352LGncgC4TwgkA4Ajr+YIcDcmzJLIZP/Irt8CwR6aYZl
 37ZN45vJ7OZVpupyVr3C/cYEu0KhAg==
 =vdZ6
 -----END PGP SIGNATURE-----

Merge tag 'drm-misc-next-2021-06-01' of git://anongit.freedesktop.org/drm/drm-misc into drm-next

drm-misc-next for 5.14:

UAPI Changes:

 * Use DRM driver names for fbdev

Cross-subsystem Changes:

Core Changes:

 * Fix leaked DMA handles

 * Improve documentation around DRM_CLIENT_CAP_*

 * Cleanups

 * dp_mst: Use kHz as link-rate unit during init

 * fourcc: Remove drm_gem_format_name() and drm_format_name_buf

 * gem-cma: Fix mmap for buffers with write combining

 * ttm: Don't override pre-set vm_ops; ttm_bo_mmap() removal and cleanups

Driver Changes:

 * drm/amdgpu: Fix hot unplug during suspend; Implement mmap as GEM object
   function; Use %p4cc format-string modifier; Cleanups

 * drm/bridge: Cdns: Fix PM reference leak, Cleanups; Lt8912b: Fix Coccinelle
   warnings; Fix Kconfig dependencies; Fixes and cleanups

 * drm/hisilicon/kirin: Cleanups

 * drm/nouveau: Implement mmap as GEM object function

 * drm/radeon: Implement mmap as GEM object function

 * drm/rockchip: Remove generic drivers during init; Add scaling for RK3036
   win1; Fix missing registers for RK3066 and 3188; Add alpha support for
   RK3036, RK3066, RK3126 and RK3188; Fixes and cleanups

 * drm/simpledrm: Use %p4cc: format-string modifier

 * drm/vmwgfx: Cleanups

 * fbdev/matrox: Use modern module_init()

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Thomas Zimmermann <tzimmermann@suse.de>
Link: https://patchwork.freedesktop.org/patch/msgid/YLZOKiYE6XFmE/MH@linux-uq9g.fritz.box
2021-06-02 12:40:26 +10:00
Nirmoy Das
57738ae40f drm/amd/display: do not dereference on NULL
WARN_ON() will not kill the process so the second WARN_ON()
will try to dereference NULL pointer.

Signed-off-by: Nirmoy Das <nirmoy.das@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:36:57 -04:00
Sathishkumar S
16eb48c62b drm/amdgpu: support atcs method powershift (v4)
add support to handle ATCS method for power shift control.
used to communicate dGPU device state to SBIOS.

V2: use defined acpi func for checking psc support (Lijo)
    fix alignment (Shashank)
V3: rebased on unified ATCS handling (Alex)
V4: rebased on ATPX/ATCS structures global (Alex)

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:36:48 -04:00
Sathishkumar S
8a81028b4f drm/amd/pm: use attr_update if the attr has it
use attr_update if its available as part of the attribute.
default_attr_update was used even if attr->attr_update is true.

Signed-off-by: Sathishkumar S <sathishkumar.sundararaju@amd.com>
Reviewed-by: Shashank Sharma <Shashank.sharma@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:36:40 -04:00
Shiwu Zhang
3c609c8b1f drm/amdgpu: free the metadata buffer for sg type BOs as well
Since both sg and device type BOs have metadata buffer, free
the buffer in both cases when to destroy BOs

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Acked-by: Nirmoy Das <Nirmoy.das@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:36:23 -04:00
Shiwu Zhang
eba9852372 drm/amdgpu: fix metadata_size for ubo ioctl queries
Although the kfd_ioctl_get_dmabuf_info() still fail it will indicate
the caller right metadat_size useful for the same kfd ioctl next time.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Nirmoy Das <nirmoy.das@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2021-06-01 22:35:50 -04:00
Stephen Boyd
6435c49913 clk: renesas: Updates for v5.14
- Add support for CPU core clock boost modes on R-Car Gen3,
   - Add ISPCS (Image Signal Processor) clocks on R-Car V3U,
   - Switch SH/R-Mobile and R-Car "DIV6" clocks to .determine_rate(),
     and improve support for multiple parents,
   - Switch RZ/N1 divider clocks to .determine_rate(),
   - Add ZA2 (Audio Clock Generator) clock on R-Car D3,
   - Minor fixes and improvements.
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYIAB0WIQQ9qaHoIs/1I4cXmEiKwlD9ZEnxcAUCYLCsnwAKCRCKwlD9ZEnx
 cLEOAQDRT3T1QFmyHD6q1lPpLUIrsp2JlWcCpNql9MRqtlttBwEAwGnDVTToYsTN
 EHtMAUanN5H2Ofvz2G7XWNwnF/hO6gA=
 =pLVr
 -----END PGP SIGNATURE-----

Merge tag 'renesas-clk-for-v5.14-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers into clk-renesas

Pull Renesas clk driver updates from Geert Uytterhoeven:

 - Add support for CPU core clock boost modes on R-Car Gen3
 - Add ISPCS (Image Signal Processor) clocks on R-Car V3U
 - Switch SH/R-Mobile and R-Car "DIV6" clocks to .determine_rate()
   and improve support for multiple parents
 - Switch RZ/N1 divider clocks to .determine_rate()
 - Add ZA2 (Audio Clock Generator) clock on R-Car D3
 - Minor fixes and improvements

* tag 'renesas-clk-for-v5.14-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-drivers:
  clk: renesas: r8a77995: Add ZA2 clock
  clk: renesas: cpg-mssr: Make srstclr[] comment block consistent
  clk: renesas: cpg-mssr: Remove unused [RM]MSTPCR() definitions
  clk: renesas: r9a06g032: Switch to .determine_rate()
  clk: renesas: div6: Implement range checking
  clk: renesas: div6: Consider all parents for requested rate
  clk: renesas: div6: Switch to .determine_rate()
  clk: renesas: div6: Simplify src mask handling
  clk: renesas: div6: Use clamp() instead of clamp_t()
  clk: renesas: rcar-usb2-clock-sel: Fix error handling in .probe()
  clk: renesas: r8a779a0: Add ISPCS clocks
  clk: renesas: rcar-gen3: Add boost support to Z clocks
  clk: renesas: rcar-gen3: Add custom clock for PLLs
  clk: renesas: rcar-gen3: Increase Z clock accuracy
  clk: renesas: rcar-gen3: Grammar s/dependent of/dependent on/
  clk: renesas: rcar-gen3: Remove superfluous masking in cpg_z_clk_set_rate()
  clk: renesas: rcar-gen3: Make cpg_z_clk.mask u32
  clk: renesas: rcar-gen3: Update Z clock rate formula in comments
2021-06-01 19:01:21 -07:00
Yevgeny Kliteynik
216214c64a net/mlx5: DR, Create multi-destination flow table with level less than 64
Flow table that contains flow pointing to multiple flow tables or multiple
TIRs must have a level lower than 64. In our case it applies to muli-
destination flow table.
Fix the level of the created table to comply with HW Spec definitions, and
still make sure that its level lower than SW-owned tables, so that it
would be possible to point from the multi-destination FW table to SW
tables.

Fixes: 34583beea4 ("net/mlx5: DR, Create multi-destination table for SW-steering use")
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:21 -07:00
Aya Levin
5349cbba75 net/mlx5e: Fix conflict with HW TS and CQE compression
When a driver's profile doesn't support a dedicated PTP-RQ,
configuration of CQE compression while HW TS is configured should fail.

Fixes: 885b8cfb16 ("net/mlx5e: Update ethtool setting of CQE compression")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:21 -07:00
Aya Levin
256f79d13c net/mlx5e: Fix HW TS with CQE compression according to profile
When the driver's profile doesn't support a dedicated PTP-RQ, the PTP
accuracy of HW TS is affected by the CQE compression. In this case,
turn off CQE compression. Otherwise, the driver crashes:

BUG: kernel NULL pointer dereference, address:0000000000000018
...
...
RIP: 0010:mlx5e_ptp_rx_set_fs+0x25/0x1a0 [mlx5_core]
...
...
Call Trace:
 mlx5e_ptp_activate_channel+0xb2/0xf0 [mlx5_core]
 mlx5e_activate_priv_channels+0x3b9/0x8c0 [mlx5_core]
 ? __mutex_unlock_slowpath+0x45/0x2a0
 ? mlx5e_refresh_tirs+0x151/0x1e0 [mlx5_core]
 mlx5e_switch_priv_channels+0x1cd/0x2d0 [mlx5_core]
 ? mlx5e_xdp_allowed+0x150/0x150 [mlx5_core]
 mlx5e_safe_switch_params+0x118/0x3c0 [mlx5_core]
 ? __mutex_lock+0x6e/0x8e0
 ? mlx5e_hwstamp_set+0xa9/0x300 [mlx5_core]
 mlx5e_hwstamp_set+0x194/0x300 [mlx5_core]
 ? dev_ioctl+0x9b/0x3d0
 mlx5i_ioctl+0x37/0x60 [mlx5_core]
 mlx5i_pkey_ioctl+0x12/0x20 [mlx5_core]
 dev_ioctl+0xa9/0x3d0
 sock_ioctl+0x268/0x420
 __x64_sys_ioctl+0x3d8/0x790
 ? lockdep_hardirqs_on_prepare+0xe4/0x190
 do_syscall_64+0x2d/0x40
entry_SYSCALL_64_after_hwframe+0x44/0xae

Fixes: 960fbfe222 ("net/mlx5e: Allow coexistence of CQE compression and HW TS PTP")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:21 -07:00
Roi Dayan
2a2c84facd net/mlx5e: Fix adding encap rules to slow path
On some devices the ignore flow level cap is not supported and we
shouldn't use it. Setting the dest ft with mlx5_chains_get_tc_end_ft()
already gives the correct end ft if ignore flow level cap is supported
or not.

Fixes: 39ac237ce0 ("net/mlx5: E-Switch, Refactor chains and priorities")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Paul Blakey <paulb@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:20 -07:00
Roi Dayan
afe93f71b5 net/mlx5e: Check for needed capability for cvlan matching
If not supported show an error and return instead of trying to offload
to the hardware and fail.

Fixes: 699e96ddf4 ("net/mlx5e: Support offloading tc double vlan headers match")
Reported-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:20 -07:00
Moshe Shemesh
5940e64281 net/mlx5: Check firmware sync reset requested is set before trying to abort it
In case driver sent NACK to firmware on sync reset request, it will get
sync reset abort event while it didn't set sync reset requested mode.
Thus, on abort sync reset event handler, driver should check reset
requested is set before trying to stop sync reset poll.

Fixes: 7dd6df329d ("net/mlx5: Handle sync reset abort event")
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:20 -07:00
Roi Dayan
b38742e411 net/mlx5e: Disable TLS offload for uplink representor
TLS offload is not supported in switchdev mode.

Fixes: 7a9fb35e8c ("net/mlx5e: Do not reload ethernet ports when changing eswitch mode")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:19 -07:00
Aya Levin
d8ec92005f net/mlx5e: Fix incompatible casting
Device supports setting of a single fec mode at a time, enforce this
by bitmap_weight == 1. Input from fec command is in u32, avoid cast to
unsigned long and use bitmap_from_arr32 to populate bitmap safely.

Fixes: 4bd9d5070b ("net/mlx5e: Enforce setting of a single FEC mode")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
2021-06-01 18:30:19 -07:00
Peter Collingbourne
419b3ab698 clk: versatile: remove dependency on ARCH_*
It is now possible to build a modular kernel for vexpress by
not setting CONFIG_ARCH_VEXPRESS=y and instead setting =m on the
drivers that it normally implies. This is with the exception of
CLK_VEXPRESS_OSC which is currently hidden behind a dependency on
one of several ARCH_* variables. Remove that dependency so that
CLK_VEXPRESS_OSC may be enabled without it.

Signed-off-by: Peter Collingbourne <pcc@google.com>
Link: https://linux-review.googlesource.com/id/I435a21e2e5f6187db54f4ef2079b60028ab2ea69
Signed-off-by: Lee Jones <lee.jones@linaro.org>
Link: https://lore.kernel.org/r/20210520161702.3746174-1-lee.jones@linaro.org
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Stephen Boyd <sboyd@kernel.org>
2021-06-01 18:18:17 -07:00
Dave Chinner
509201163f xfs: remove xfs_perag_t
Almost unused, gets rid of another typedef.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:51 +10:00
Dave Chinner
f40aadb2bb xfs: use perag through unlink processing
Unlinked lists are held in the perag, and freeing of inodes needs to
be passed a perag, too, so look up the perag early in the unlink
processing and use it throughout.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Brian Foster <bfoster@redhat.com>
2021-06-02 10:48:51 +10:00
Dave Chinner
8237fbf53d xfs: clean up and simplify xfs_dialloc()
Because it's a mess.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
309161f660 xfs: inode allocation can use a single perag instance
Now that we've internalised the two-phase inode allocation, we can
now easily make the AG selection and allocation atomic from the
perspective of a single perag context. This will ensure AGs going
offline/away cannot occur between the selection and allocation
steps.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
b652afd937 xfs: get rid of xfs_dir_ialloc()
This is just a simple wrapper around the per-ag inode allocation
that doesn't need to exist. The internal mechanism to select and
allocate within an AG does not need to be exposed outside
xfs_ialloc.c, and it being exposed simply makes it harder to follow
the code and simplify it.

This is simplified by internalising xf_dialloc_select_ag() and
xfs_dialloc_ag() into a single xfs_dialloc() function and then
xfs_dir_ialloc() can go away.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
89b1f55a29 xfs: collapse AG selection for inode allocation
xfs_dialloc_select_ag() does a lot of repetitive work. It first
calls xfs_ialloc_ag_select() to select the AG to start allocation
attempts in, which can do up to two entire loops across the perags
that inodes can be allocated in. This is simply checking if there is
spce available to allocate inodes in an AG, and it returns when it
finds the first candidate AG.

xfs_dialloc_select_ag() then does it's own iterative walk across
all the perags locking the AGIs and trying to allocate inodes from
the locked AG. It also doesn't limit the search to mp->m_maxagi,
so it will walk all AGs whether they can allocate inodes or not.

Hence if we are really low on inodes, we could do almost 3 entire
walks across the whole perag range before we find an allocation
group we can allocate inodes in or report ENOSPC.

Because xfs_ialloc_ag_select() returns on the first candidate AG it
finds, we can simply do these checks directly in
xfs_dialloc_select_ag() before we lock and try to allocate inodes.
This reduces the inode allocation pass down to 2 perag sweeps at
most - one for aligned inode cluster allocation and if we can't
allocate full, aligned inode clusters anywhere we'll do another pass
trying to do sparse inode cluster allocation.

This also removes a big chunk of duplicate code.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
4268547305 xfs: simplify xfs_dialloc_select_ag() return values
The only caller of xfs_dialloc_select_ag() will always return
-ENOSPC to it's caller if the agbp returned from
xfs_dialloc_select_ag() is NULL. IOWs, failure to find a candidate
AGI we can allocate inodes from is always an ENOSPC condition, so
move this logic up into xfs_dialloc_select_ag() so we can simplify
the return logic in this function.

xfs_dialloc_select_ag() now only ever returns 0 with a locked
agbp, or an error with no agbp.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
50f02fe333 xfs: remove agno from btree cursor
Now that everything passes a perag, the agno is not needed anymore.
Convert all the users to use pag->pag_agno instead and remove the
agno from the cursor. This was largely done as an automated search
and replace.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
7b13c51551 xfs: use perag for ialloc btree cursors
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
289d38d22c xfs: convert allocbt cursors to use perags
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
a81a06211f xfs: convert refcount btree cursor to use perags
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
fa9c3c1973 xfs: convert rmap btree cursor to using a perag
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
be9fb17d88 xfs: add a perag to the btree cursor
Which will eventually completely replace the agno in it.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Brian Foster <bfoster@redhat.com>
2021-06-02 10:48:24 +10:00
Dave Chinner
58d43a7e32 xfs: pass perags around in fsmap data dev functions
Needs a [from, to] ranged AG walk, and the perag to be stuffed into
the info structure for callouts to use.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
30933120ad xfs: push perags through the ag reservation callouts
We currently pass an agno from the AG reservation functions to the
individual feature accounting functions, which in future may have to
do perag lookups to access per-AG state. Instead, pre-emptively
plumb the perag through from the highest AG reservation layer to the
feature callouts so they won't have to look it up again.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Brian Foster <bfoster@redhat.com>
2021-06-02 10:48:24 +10:00
Dave Chinner
45d0662117 xfs: pass perags through to the busy extent code
All of the callers of the busy extent API either have perag
references available to use so we can pass a perag to the busy
extent functions rather than having them have to do unnecessary
lookups.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
7f8d3b3ca6 xfs: convert secondary superblock walk to use perags
Clean up the last external manual AG walk.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
6f4118fc64 xfs: convert xfs_iwalk to use perag references
Rather than manually walking the ags and passing agnunbers around,
pass the perag for the AG we are currently working on around in the
iwalk structure.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
934933c3ee xfs: convert raw ag walks to use for_each_perag
Convert the raw walks to an iterator, pulling the current AG out of
pag->pag_agno instead of the loop iterator variable.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
f250eedcf7 xfs: make for_each_perag... a first class citizen
for_each_perag_tag() is defined in xfs_icache.c for local use.
Promote this to xfs_ag.h and define equivalent iteration functions
so that we can use them to iterate AGs instead to replace open coded
perag walks and perag lookups.

We also convert as many of the straight forward open coded AG walks
to use these iterators as possible. Anything that is not a direct
conversion to an iterator is ignored and will be updated in future
commits.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
07b6403a68 xfs: move perag structure and setup to libxfs/xfs_ag.[ch]
Move the xfs_perag infrastructure to the libxfs files that contain
all the per AG infrastructure. This helps set up for passing perags
around all the code instead of bare agnos with minimal extra
includes for existing files.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
61aa005a5b xfs: prepare for moving perag definitions and support to libxfs
The perag structures really need to be defined with the rest of the
AG support infrastructure. The struct xfs_perag and init/teardown
has been placed in xfs_mount.[ch] because there are differences in
the structure between kernel and userspace. Mainly that userspace
doesn't have a lot of the internal stuff that the kernel has for
caches and discard and other such structures.

However, it makes more sense to move this to libxfs than to keep
this separation because we are now moving to use struct perags
everywhere in the code instead of passing raw agnumber_t values
about. Hence we shoudl really move the support infrastructure to
libxfs/xfs_ag.[ch].

To do this without breaking userspace, first we need to rearrange
the structures and code so that all the kernel specific code is
located together. This makes it simple for userspace to ifdef out
the all the parts it does not need, minimising the code differences
between kernel and userspace. The next commit will do the move...

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00
Dave Chinner
9bbafc7191 xfs: move xfs_perag_get/put to xfs_ag.[ch]
They are AG functions, not superblock functions, so move them to the
appropriate location.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
2021-06-02 10:48:24 +10:00