If a writeback of the superblock fails with an I/O error, the buffer
is marked not uptodate. However, this can cause a WARN_ON to trigger
when we attempt to write superblock a second time. (Which might
succeed this time, for cerrtain types of block devices such as iSCSI
devices over a flaky network.)
Try to detect this case in flush_stashed_error_work(), and also change
__ext4_handle_dirty_metadata() so we always set the uptodate flag, not
just in the nojournal case.
Before this commit, this problem can be repliciated via:
1. dmsetup create dust1 --table '0 2097152 dust /dev/sdc 0 4096'
2. mount /dev/mapper/dust1 /home/test
3. dmsetup message dust1 0 addbadblock 0 10
4. cd /home/test
5. echo "XXXXXXX" > t
After a few seconds, we got following warning:
[ 80.654487] end_buffer_async_write: bh=0xffff88842f18bdd0
[ 80.656134] Buffer I/O error on dev dm-0, logical block 0, lost async page write
[ 85.774450] EXT4-fs error (device dm-0): ext4_check_bdev_write_error:193: comm kworker/u16:8: Error while async write back metadata
[ 91.415513] mark_buffer_dirty: bh=0xffff88842f18bdd0
[ 91.417038] ------------[ cut here ]------------
[ 91.418450] WARNING: CPU: 1 PID: 1944 at fs/buffer.c:1092 mark_buffer_dirty.cold+0x1c/0x5e
[ 91.440322] Call Trace:
[ 91.440652] __jbd2_journal_temp_unlink_buffer+0x135/0x220
[ 91.441354] __jbd2_journal_unfile_buffer+0x24/0x90
[ 91.441981] __jbd2_journal_refile_buffer+0x134/0x1d0
[ 91.442628] jbd2_journal_commit_transaction+0x249a/0x3240
[ 91.443336] ? put_prev_entity+0x2a/0x200
[ 91.443856] ? kjournald2+0x12e/0x510
[ 91.444324] kjournald2+0x12e/0x510
[ 91.444773] ? woken_wake_function+0x30/0x30
[ 91.445326] kthread+0x150/0x1b0
[ 91.445739] ? commit_timeout+0x20/0x20
[ 91.446258] ? kthread_flush_worker+0xb0/0xb0
[ 91.446818] ret_from_fork+0x1f/0x30
[ 91.447293] ---[ end trace 66f0b6bf3d1abade ]---
Signed-off-by: Ye Bin <yebin10@huawei.com>
Link: https://lore.kernel.org/r/20210615090537.3423231-1-yebin10@huawei.com
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
An IRQ handler may be called at any time after it is registered, so
anything it relies on must be ready before registration.
rockchip_pcie_subsys_irq_handler() and rockchip_pcie_client_irq_handler()
read registers in the PCIe controller, but we registered them before
turning on clocks to the controller. If either is called before the clocks
are turned on, the register reads fail and the machine hangs.
Similarly, rockchip_pcie_legacy_int_handler() uses rockchip->irq_domain,
but we installed it before initializing irq_domain.
Register IRQ handlers after their data structures are initialized and
clocks are enabled.
Found by enabling CONFIG_DEBUG_SHIRQ, which calls the IRQ handler when it
is being unregistered. An error during the probe path might cause this
unregistration and IRQ handler execution before the device or data
structure init has finished.
[bhelgaas: commit log]
Link: https://lore.kernel.org/r/20210608080409.1729276-1-javierm@redhat.com
Reported-by: Peter Robinson <pbrobinson@gmail.com>
Tested-by: Peter Robinson <pbrobinson@gmail.com>
Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Shawn Lin <shawn.lin@rock-chips.com>
Commit 9b69d48c75 ("powerpc/64e: remove implicit soft-masking and
interrupt exit restart logic") limited the implicit soft masking and
restart logic to 64-bit Book3S only. However we are still building
restart_table.c for all 64-bit, ie. Book3E also.
There's no need to build it for 64e, and it also causes missing
prototype warnings for 64e builds, because the prototype is already
behind an #ifdef PPC_BOOK3S_64.
Fixes: 9b69d48c75 ("powerpc/64e: remove implicit soft-masking and interrupt exit restart logic")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210701125026.292224-1-mpe@ellerman.id.au
Pull ARM cpufreq updates for v5.14-rc1 from Viresh Kumar:
"- Add frequency invariance support for CPPC driver again and related
fixes/changes."
- Minor changes/cleanups for Meditak driver (Fabien Parent and Seiya
Wang), Qcom platform (Sibi Sankar), and SCMI driver (Christophe
JAILLET).
- New bindings for generic performance domains (Sudeep Holla).
- Rename black/white-lists (Viresh Kumar)."
* 'cpufreq/arm/linux-next' of git://git.kernel.org/pub/scm/linux/kernel/git/vireshk/pm:
cpufreq: CPPC: Add support for frequency invariance
arch_topology: Avoid use-after-free for scale_freq_data
cpufreq: CPPC: Pass structure instance by reference
cpufreq: CPPC: Fix potential memleak in cppc_cpufreq_cpu_init
dt-bindings: cpufreq: update cpu type and clock name for MT8173 SoC
clk: mediatek: remove deprecated CLK_INFRA_CA57SEL for MT8173 SoC
cpufreq: dt: Rename black/white-lists
cpufreq: scmi: Fix an error message
cpufreq: mediatek: add support for mt8365
dt-bindings: dvfs: Add support for generic performance domains
cpufreq: blacklist SC7280 in cpufreq-dt-platdev
ASoC: Updates for v5.14
This release sees a nice new feature in the core from Morimoto-san,
support for automatic negotiation of DAI formats between the components
on the link. Otherwise the big highlight was the merging of the Tegra
machine drivers into a single driver avoiding a bunch of duplication.
- Support for automatic negotiation of DAI formats.
- Accessory detection support for several Qualcomm parts.
- Support for IEC958 control with hdmi-codec.
- Merging of Tegra machine drivers into a single driver.
- Support for AmLogic SM1 TOACODEC, Intel AlderLake-M, several NXP
i.MX8 variants, NXP TFA1 and TDF9897, Rockchip RK817, Qualcomm
Quinary MI2S, Texas Instruments TAS2505
This contains both the short-term fix for the W+X boot mappings and the
larger cleanup.
* riscv-wx-mappings:
riscv: Map the kernel with correct permissions the first time
riscv: Introduce set_kernel_memory helper
riscv: Simplify xip and !xip kernel address conversion macros
riscv: Remove CONFIG_PHYS_RAM_BASE_FIXED
riscv: mm: Fix W+X mappings at boot
For 64-bit kernels, we map all the kernel with write and execute
permissions and afterwards remove writability from text and executability
from data.
For 32-bit kernels, the kernel mapping resides in the linear mapping, so we
map all the linear mapping as writable and executable and afterwards we
remove those properties for unused memory and kernel mapping as
described above.
Change this behavior to directly map the kernel with correct permissions
and avoid going through the whole mapping to fix the permissions.
At the same time, this fixes an issue introduced by commit 2bfc6cd81b
("riscv: Move kernel mapping outside of linear mapping") as reported
here https://github.com/starfive-tech/linux/issues/17.
Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
Reviewed-by: Anup Patel <anup@brainfault.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Each zone-device page holds a reference to the SVM BO that manages its
backing storage. This is necessary to correctly hold on to the BO in
case zone_device pages are shared with a child-process.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This is for debug purposes only.
It conditionally generates partial migrations to test mixed
CPU/GPU memory domain pages in a prange easily.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Migration skipped for pages that are already in VRAM
domain. These could be the result of previous partial
migrations to SYS RAM, and prefetch back to VRAM.
Ex. Coherent pages in VRAM that were not written/invalidated after
a copy-on-write.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Invalid pages can be the result of pages that have been migrated
already due to copy-on-write procedure or pages that were never
migrated to VRAM in first place. This is not an issue anymore,
as pranges now support mixed memory domains (CPU/GPU).
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
svm ranges can have mixed pages from device or system memory.
A good example is, after a prange has been allocated in VRAM and a
copy-on-write is triggered by a fork. This invalidates some pages
inside the prange. Endding up in mixed pages.
[How]
By classifying each page inside a prange, based on its type. Device or
system memory, during dma mapping call. If page corresponds
to VRAM domain, a flag is set to its dma_addr entry for each GPU.
Then, at the GPU page table mapping. All group of contiguous pages within
the same type are mapped with their proper pte flags.
v2:
Instead of using ttm_res to calculate vram pfns in the svm_range. It is now
done by setting the vram real physical address into drm_addr array.
This makes more flexible VRAM management, plus removes the need to have
a BO reference in the svm_range.
v3:
Remove mapping member from svm_range
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Now that prange could have mixed domains (VRAM or SYSRAM),
actual_loc nor svm_bo can not be used to check its current
domain and eventually get its pfns to map them in GPU.
Instead, pfns from both domains, are now obtained from
hmm_range_fault through amdgpu_hmm_range_get_pages
call. This is done everytime a GPU map occur.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Get the proper owner reference for amdgpu_hmm_range_get_pages function.
This is useful for partial migrations. To avoid migrating back to
system memory, VRAM pages, that are accessible by all devices in the
same memory domain.
Ex. multiple devices in the same hive.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
svm_range_prefault is called right before migrations to VRAM,
to make sure pages are resident in system memory before the migration.
With partial migrations, this reference is used by hmm range get pages
to avoid migrating pages that are already in the same VRAM domain.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The parameter is used in the dev_private_owner to decide if device
pages in the range require to be migrated back to system memory, based
if they are or not in the same memory domain.
In this case, this reference could come from the same memory domain
with devices connected to the same hive.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
GPUs in the same XGMI hive have direct access to all
members'VRAM. When mapping memory to a GPU, we don't need
hmm_range_fault to fault device-private pages in the same
hive back to the host. Identifying the page owner as the hive,
rather than the individual GPU, accomplishes this.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
During GPU page table invalidation with xnack off, new ranges
split may occur concurrently in the same prange. Creating a new
child per split. Each child should also increment its
invalid counter, to assure GPU page table updates in these
ranges.
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why & How]
Extend existing support for DCN2.1 DMUB diagnostic logging to
DCN3.1 so we can collect useful information if the DMUB hangs.
Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Navi series GPUs have 2 SIMDs per CU (and then 2 CUs per WGP).
The NV enum headers incorrectly listed this as 4, which later meant
we were incorrectly reporting the number of SIMDs in the HSA
topology. This could cause problems down the line for user-space
applications that want to launch a fixed amount of work to each
SIMD.
Signed-off-by: Joseph Greathouse <Joseph.Greathouse@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
The documentation around PrepareMp1ForUnload message says that
anything sent to SMU after this command would be stalled as the
PMFW would not be in a state to take further job requests.
Technically this is right in case of S3 scenario. But, this might
not be the case during s0ix as the PMC driver would be the last
to send the SMU on the OS_HINT. If SMU gets a PrepareMp1ForUnload
message before the OS_HINT, this would stall the entire S0ix process.
Results show that, this message to SMU is not required during S0ix
and hence skip it.
Reviewed-by: Prike Liang <Prike.Liang@amd.com>
Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
Acked-by: Huang Rui <ray.huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In some asics, we need to adjust the behavior according to the apu flags
at very early stage.
Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Aaron Liu <aaron.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Setting CONFIG_FRAME_WARN=0 should disable 'stack frame larger than'
warnings. This is useful for example in KASAN builds. Make the dml
Makefile respect this config.
Fixes the following build warnings with CONFIG_KASAN=y and
CONFIG_FRAME_WARN=0:
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn30/display_mode_vba_30.c:3642:6:
warning: stack frame size of 2216 bytes in function
'dml30_ModeSupportAndSystemConfigurationFull' [-Wframe-larger-than=]
drivers/gpu/drm/amd/amdgpu/../display/dc/dml/dcn31/display_mode_vba_31.c:3957:6:
warning: stack frame size of 2568 bytes in function
'dml31_ModeSupportAndSystemConfigurationFull' [-Wframe-larger-than=]
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Reka Norman <rekanorman@google.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Also copy over the part that makes old gcc handling cross-platform.
Fixes: df7a1658f2 ("drm/amdgpu/dc: fix DCN3.1 Makefile for PPC64")
Fixes: 926d6972ef ("drm/amd/display: Add DCN3.1 blocks to the DC Makefile")
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Michal Suchanek <msuchanek@suse.de>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
On the Loongson64 platform used with Radeon GPU, shutdown or reboot failed
when console=tty is in the boot cmdline.
radeon_suspend_kms() puts the hw in the suspend state, especially set fb
state as FBINFO_STATE_SUSPENDED:
if (fbcon) {
console_lock();
radeon_fbdev_set_suspend(rdev, 1);
console_unlock();
}
Then avoid to do any more fb operations in the related functions:
if (p->state != FBINFO_STATE_RUNNING)
return;
So call radeon_suspend_kms() in radeon_pci_shutdown() for Loongson64 to fix
this issue, it looks like some kind of workaround like powerpc.
Co-developed-by: Jianmin Lv <lvjianmin@loongson.cn>
Signed-off-by: Jianmin Lv <lvjianmin@loongson.cn>
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
The ttm caching flags (ttm_cached, ttm_write_combined etc) are
used to determine a buffer object's mapping attributes in both
CPU page table and GPU page table (when that buffer is also
accessed by GPU). Currently the ttm caching flags are set in
function amdgpu_ttm_io_mem_reserve which is called during
DRM_AMDGPU_GEM_MMAP ioctl. This has a problem since the GPU
mapping of the buffer object (ioctl DRM_AMDGPU_GEM_VA) can
happen earlier than the mmap time, thus the GPU page table
update code can't pick up the right ttm caching flags to
decide the right GPU page table attributes.
This patch moves the ttm caching flags setting to function
amdgpu_vram_mgr_new - this function is called during the
first step of a buffer object create (eg, DRM_AMDGPU_GEM_CREATE)
so the later both CPU and GPU mapping function calls will
pick up this flag for CPU/GPU page table set up.
v2: rebase (Alex)
Signed-off-by: Oak Zeng <Oak.Zeng@amd.com>
Suggested-by: Christian Koenig <Christian.Koenig@amd.com>
Reviewed-by: Christian Koenig <Christian.Koenig@amd.com>
Reviewed-by: Feifei Xu <Feifei.Xu@amd.com>
Tested-by: Po Huang <Po.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
During GPU reset, when receiving a DMCUB OUTBUX0 interrupt,
DAL code will set it to be OUTBOX interrupt and sets hw interrupt.
However, OUTBOX interrupt is not registered yet, so a NULL pointer
access will be executed.
Call Trace:
dal_irq_service_set+0x30/0x90 [amdgpu]
dc_interrupt_set+0x24/0x30 [amdgpu]
amdgpu_dm_set_dmub_outbox_irq_state+0x22/0x30 [amdgpu]
amdgpu_irq_update+0x77/0xa0 [amdgpu]
amdgpu_irq_gpu_reset_resume_helper+0x67/0xa0 [amdgpu]
amdgpu_do_asic_reset+0x219/0x260 [amdgpu]
amdgpu_device_gpu_recover.cold+0x8c5/0xb64 [amdgpu]
amdgpu_debugfs_gpu_recover_show+0x2c/0x60 [amdgpu]
seq_read_iter+0xc2/0x450
? do_anonymous_page+0x22c/0x3b0
seq_read+0xf9/0x140
full_proxy_read+0x5c/0x90
vfs_read+0xaa/0x190
ksys_read+0x67/0xe0
__x64_sys_read+0x1a/0x20
Fixes: effbf6ca7e ("drm/amdgpu/display: remove an old DCN3 guard")
Signed-off-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-and-tested-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Without driver loaded, SDMA0_UTCL1_PAGE.TMZ_ENABLE is set to 1
by default for all asic. On Raven/Renoir, the sdma goldsetting
changes SDMA0_UTCL1_PAGE.TMZ_ENABLE to 0.
This patch restores SDMA0_UTCL1_PAGE.TMZ_ENABLE to 1.
Signed-off-by: Aaron Liu <aaron.liu@amd.com>
Acked-by: Luben Tuikov <luben.tuikov@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
This helper should be used for setting permissions to the kernel
mapping as it takes pointers as arguments and then avoids explicit cast
to unsigned long needed for the set_memory_* API.
Suggested-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
Reviewed-by: Anup Patel <anup@brainfault.org>
Reviewed-by: Atish Patra <atish.patra@wdc.com>
Reviewed-by: Jisheng Zhang <jszhang@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Add architecture specific implementation details for KFENCE and enable
KFENCE for the riscv64 architecture. In particular, this implements the
required interface in <asm/kfence.h>.
KFENCE requires that attributes for pages from its memory pool can
individually be set. Therefore, force the kfence pool to be mapped at
page granularity.
Testing this patch using the testcases in kfence_test.c and all passed.
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Acked-by: Marco Elver <elver@google.com>
Reviewed-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Implement optimized version of the tlb flushing routines for systems
using ASIDs. These are behind the use_asid_allocator static branch to
not affect existing systems not using ASIDs.
Signed-off-by: Guo Ren <guoren@linux.alibaba.com>
[hch: rebased on top of previous cleanups, use the same algorithm as
the non-ASID based code for local vs global flushes, keep functions
as local as possible]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Guo Ren <guoren@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Move the call mm_cpumask from the callers into __sbi_tlb_flush_range to
reduce a bit of duplicate code and prepare for future changes.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Tested-by: Guo Ren <guoren@kernel.org>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
Currently, memory-hotplug code takes zone's span_writelock and pgdat's
resize_lock when resizing the node/zone's spanned pages via
{move_pfn_range_to_zone(),remove_pfn_range_from_zone()} and when resizing
node and zone's present pages via adjust_present_page_count().
These locks are also taken during the initialization of the system at boot
time, where it protects parallel struct page initialization, but they
should not really be needed in memory-hotplug where all operations are a)
synchronized on device level and b) serialized by the mem_hotplug_lock
lock.
[akpm@linux-foundation.org: remove now-unused locals]
Link: https://lkml.kernel.org/r/20210531093958.15021-1-osalvador@suse.de
Signed-off-by: Oscar Salvador <osalvador@suse.de>
Acked-by: David Hildenbrand <david@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
