powerpc/vmemmap: Fix memory leak with vmemmap list allocation failures.
Subsystems (38)
Sasha Levin
maintainerPATCH 1/147 powerpc/vmemmap: Fix memory leak with vmemmap list allocation failures.
From: "Aneesh Kumar K.V" <[email protected]>
[ Upstream commit ccaea15296f9773abd43aaa17ee4b88848e4a505 ]
If we fail to allocate vmemmap list, we don't keep track of allocated vmemmap block buf. Hence on section deactivate we skip vmemmap block buf free. This results in memory leak.
--- arch/powerpc/mm/init_64.c | 35 ++++++++++++++++++++++++++++------- 1 file changed, 28 insertions(+), 7 deletions(-)
arch/powerpc/mm/init_64.c +28 -7
@@ -162,16 +162,16 @@ static __meminit struct vmemmap_backing * vmemmap_list_alloc(int node)return next++; } -static __meminit void vmemmap_list_populate (unsigned long phys, -unsigned long start, -int node) +static __meminit int vmemmap_list_populate (unsigned long phys, +unsigned long start, +int node) {struct vmemmap_backing *vmem_back; vmem_back =vmemmap_list_alloc (node);if (unlikely (!vmem_back)) { -WARN_ON (1 ); -return ; +pr_debug ("vmemap list allocation failed\n" ); +return -ENOMEM ; } vmem_back->phys = phys;
@@ -179,6 +179,7 @@ static __meminit void vmemmap_list_populate(unsigned long phys, vmem_back->list = vmemmap_list; vmemmap_list = vmem_back; +return 0 ; } static bool altmap_cross_boundary(struct vmem_altmap *altmap, unsigned long start,
@@ -199,6 +200,7 @@ static bool altmap_cross_boundary(struct vmem_altmap *altmap, unsigned long starint __meminit vmemmap_populate (unsigned long start,unsigned long end,int node,struct vmem_altmap *altmap) { +bool altmap_alloc;unsigned long page_size =1 << mmu_psize_defs[mmu_vmemmap_psize].shift;/* Align to the page size of the linear mapping. */
@@ -228,13 +230,32 @@ int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node, p =vmemmap_alloc_block_buf (page_size, node, altmap);if (!p)pr_debug ("altmap block allocation failed, falling back to system memory" ); +else + altmap_alloc = true; } -if (!p) +if (!p) { p =vmemmap_alloc_block_buf (page_size, node,NULL ); + altmap_alloc = false; + }if (!p)return -ENOMEM ; -vmemmap_list_populate (__pa (p), start, node); +if (vmemmap_list_populate (__pa (p), start, node)) { +/* +* If we don't populate vmemap list, we don't have +* the ability to free the allocated vmemmap +* pages in section_deactivate. Hence free them +* here. +*/ +int nr_pfns = page_size >>PAGE_SHIFT ; +unsigned long page_order =get_order (page_size); + +if (altmap_alloc) +vmem_altmap_free (altmap, nr_pfns); +else +free_pages ((unsigned long )p, page_order); +return -ENOMEM ; + }pr_debug (" * %016lx..%016lx allocated at %p\n" , start, start + page_size, p);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 2/147 powerpc/powernv/smp: Fix spurious DBG() warning
From: Oliver O'Halloran <[email protected]>
[ Upstream commit f6bac19cf65c5be21d14a0c9684c8f560f2096dd ]
When building with W=1 we get the following warning:
arch/powerpc/platforms/powernv/smp.c: In function ‘pnv_smp_cpu_kill_self’:
arch/powerpc/platforms/powernv/smp.c:276:16: error: suggest braces around
empty body in an ‘if’ statement [-Werror=empty-body]
276 | cpu, srr1);
| ^
cc1: all warnings being treated as errors
The full context is this block:
if (srr1 && !generic_check_cpu_restart(cpu))
DBG("CPU%d Unexpected exit while offline srr1=%lx!\n",
cpu, srr1);
When building with DEBUG undefined DBG() expands to nothing and GCC emits the warning due to the lack of braces around an empty statement.
--- arch/powerpc/platforms/powernv/smp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/platforms/powernv/smp.c +1 -1
@@ -43,7 +43,7 @@#include <asm/udbg.h> #define DBG(fmt...) udbg_printf(fmt)#else -#define DBG(fmt...) +#define DBG(fmt...) do { } while (0)#endif static void pnv_smp_setup_cpu (int cpu)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 3/147 RDMA/core: Change how failing destroy is handled during uobj abort
In reply to Sasha LevinFrom: Jason Gunthorpe <[email protected]>
[ Upstream commit f553246f7f794675da1794ae7ee07d1f35e561ae ]
Currently it triggers a WARN_ON and then goes ahead and destroys the uobject anyhow, leaking any driver memory.
The only place that leaks driver memory should be during FD close() in uverbs_destroy_ufile_hw().
Drivers are only allowed to fail destroy uobjects if they guarantee destroy will eventually succeed. uverbs_destroy_ufile_hw() provides the loop to give the driver that chance.
--- drivers/infiniband/core/rdma_core.c | 30 ++++++++++++++--------------- include/rdma/ib_verbs.h | 5 ----- 2 files changed, 15 insertions(+), 20 deletions(-)
drivers/infiniband/core/rdma_core.c +15 -15
@@ -130,17 +130,6 @@ static int uverbs_destroy_uobject(struct ib_uobject *uobj,lockdep_assert_held (&ufile->hw_destroy_rwsem);assert_uverbs_usecnt (uobj,UVERBS_LOOKUP_WRITE ); -if (reason ==RDMA_REMOVE_ABORT_HWOBJ ) { - reason =RDMA_REMOVE_ABORT ; - ret = uobj->uapi_object->type_class->destroy_hw (uobj, reason, - attrs); -/* -* Drivers are not permitted to ignore RDMA_REMOVE_ABORT, see -* ib_is_destroy_retryable, cleanup_retryable == false here. -*/ -WARN_ON (ret); - } -if (reason ==RDMA_REMOVE_ABORT ) {WARN_ON (!list_empty (&uobj->list));WARN_ON (!uobj->context);
@@ -674,11 +663,22 @@ void rdma_alloc_abort_uobject(struct ib_uobject *uobj, bool hw_obj_valid) {struct ib_uverbs_file *ufile = uobj->ufile; +int ret; + +if (hw_obj_valid) { + ret = uobj->uapi_object->type_class->destroy_hw ( + uobj,RDMA_REMOVE_ABORT , attrs); +/* +* If the driver couldn't destroy the object then go ahead and +* commit it. Leaking objects that can't be destroyed is only +* done during FD close after the driver has a few more tries to +* destroy it. +*/ +if (WARN_ON (ret)) +return rdma_alloc_commit_uobject (uobj, attrs); + } -uverbs_destroy_uobject (uobj, - hw_obj_valid ?RDMA_REMOVE_ABORT_HWOBJ : -RDMA_REMOVE_ABORT , - attrs); +uverbs_destroy_uobject (uobj,RDMA_REMOVE_ABORT , attrs);/* Matches the down_read in rdma_alloc_begin_uobject */ up_read (&ufile->hw_destroy_rwsem);
include/rdma/ib_verbs.h +0 -5
@@ -1463,11 +1463,6 @@ enum rdma_remove_reason {RDMA_REMOVE_DRIVER_REMOVE ,/* uobj is being cleaned-up before being committed */ RDMA_REMOVE_ABORT, -/* -* uobj has been fully created, with the uobj->object set, but is being -* cleaned up before being comitted -*/ - RDMA_REMOVE_ABORT_HWOBJ, }; struct ib_rdmacg_object {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 4/147 f2fs: allocate proper size memory for zstd decompress
In reply to Sasha LevinFrom: Chao Yu <[email protected]>
[ Upstream commit 0e2b7385cb59e566520cfd0a04b4b53bc9461e98 ]
As 5kft <[email protected]> reported:
kworker/u9:3: page allocation failure: order:9, mode:0x40c40(GFP_NOFS|__GFP_COMP), nodemask=(null),cpuset=/,mems_allowed=0
CPU: 3 PID: 8168 Comm: kworker/u9:3 Tainted: G C 5.8.3-sunxi #trunk Hardware name: Allwinner sun8i Family Workqueue: f2fs_post_read_wq f2fs_post_read_work [<c010d6d5>] (unwind_backtrace) from [<c0109a55>] (show_stack+0x11/0x14) [<c0109a55>] (show_stack) from [<c056d489>] (dump_stack+0x75/0x84) [<c056d489>] (dump_stack) from [<c0243b53>] (warn_alloc+0xa3/0x104) [<c0243b53>] (warn_alloc) from [<c024473b>] (__alloc_pages_nodemask+0xb87/0xc40) [<c024473b>] (__alloc_pages_nodemask) from [<c02267c5>] (kmalloc_order+0x19/0x38) [<c02267c5>] (kmalloc_order) from [<c02267fd>] (kmalloc_order_trace+0x19/0x90) [<c02267fd>] (kmalloc_order_trace) from [<c047c665>] (zstd_init_decompress_ctx+0x21/0x88) [<c047c665>] (zstd_init_decompress_ctx) from [<c047e9cf>] (f2fs_decompress_pages+0x97/0x228) [<c047e9cf>] (f2fs_decompress_pages) from [<c045d0ab>] (__read_end_io+0xfb/0x130) [<c045d0ab>] (__read_end_io) from [<c045d141>] (f2fs_post_read_work+0x61/0x84) [<c045d141>] (f2fs_post_read_work) from [<c0130b2f>] (process_one_work+0x15f/0x3b0) [<c0130b2f>] (process_one_work) from [<c0130e7b>] (worker_thread+0xfb/0x3e0) [<c0130e7b>] (worker_thread) from [<c0135c3b>] (kthread+0xeb/0x10c) [<c0135c3b>] (kthread) from [<c0100159>]
zstd may allocate large size memory for {,de}compression, it may cause file copy failure on low-end device which has very few memory.
For decompression, let's just allocate proper size memory based on current file's cluster size instead of max cluster size.
--- fs/f2fs/compress.c | 7 ++++--- fs/f2fs/f2fs.h | 2 +- 2 files changed, 5 insertions(+), 4 deletions(-)
fs/f2fs/compress.c +4 -3
@@ -382,16 +382,17 @@ static int zstd_init_decompress_ctx(struct decompress_io_ctx *dic)ZSTD_DStream *stream;void *workspace;unsigned int workspace_size; +unsigned int max_window_size = +MAX_COMPRESS_WINDOW_SIZE (dic->log_cluster_size); - workspace_size =ZSTD_DStreamWorkspaceBound (MAX_COMPRESS_WINDOW_SIZE ); + workspace_size =ZSTD_DStreamWorkspaceBound (max_window_size); workspace =f2fs_kvmalloc (F2FS_I_SB (dic->inode), workspace_size,GFP_NOFS );if (!workspace)return -ENOMEM ; - stream =ZSTD_initDStream (MAX_COMPRESS_WINDOW_SIZE , - workspace, workspace_size); + stream =ZSTD_initDStream (max_window_size, workspace, workspace_size);if (!stream) { printk_ratelimited("%sF2FS-fs (%s): %s ZSTD_initDStream failed\n", KERN_ERR, F2FS_I_SB(dic->inode)->sb->s_id,
fs/f2fs/f2fs.h +1 -1
@@ -1394,7 +1394,7 @@ struct decompress_io_ctx {#define NULL_CLUSTER ((unsigned int)(~0))#define MIN_COMPRESS_LOG_SIZE 2#define MAX_COMPRESS_LOG_SIZE 8 -#define MAX_COMPRESS_WINDOW_SIZE ((PAGE_SIZE) << MAX_COMPRESS_LOG_SIZE) +#define MAX_COMPRESS_WINDOW_SIZE (log_size) ((PAGE_SIZE) << (log_size))struct f2fs_sb_info {struct super_block *sb;/* pointer to VFS super block */
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 5/147 powerpc/watchpoint/ptrace: Fix SETHWDEBUG when CONFIG_HAVE_HW_BREAKPOINT=N
In reply to Sasha LevinFrom: Ravi Bangoria <[email protected]>
[ Upstream commit 9b6b7c680cc20971444d9f836e49fc98848bcd0a ]
When kernel is compiled with CONFIG_HAVE_HW_BREAKPOINT=N, user can still create watchpoint using PPC_PTRACE_SETHWDEBUG, with limited functionalities. But, such watchpoints are never firing because of the missing privilege settings. Fix that.
It's safe to set HW_BRK_TYPE_PRIV_ALL because we don't really leak any kernel address in signal info. Setting HW_BRK_TYPE_PRIV_ALL will also help to find scenarios when kernel accesses user memory.
--- arch/powerpc/kernel/ptrace/ptrace-noadv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/kernel/ptrace/ptrace-noadv.c +1 -1
@@ -217,7 +217,7 @@ long ppc_set_hwdebug(struct task_struct *child, struct ppc_hw_breakpoint *bp_infreturn -EIO ; brk.address =ALIGN_DOWN (bp_info->addr,HW_BREAKPOINT_SIZE ); - brk.type =HW_BRK_TYPE_TRANSLATE ; + brk.type =HW_BRK_TYPE_TRANSLATE |HW_BRK_TYPE_PRIV_ALL ; brk.len =DABR_MAX_LEN ;if (bp_info->trigger_type &PPC_BREAKPOINT_TRIGGER_READ ) brk.type |=HW_BRK_TYPE_READ ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 6/147 mm: fix exec activate_mm vs TLB shootdown and lazy tlb switching race
In reply to Sasha LevinFrom: Nicholas Piggin <[email protected]>
[ Upstream commit d53c3dfb23c45f7d4f910c3a3ca84bf0a99c6143 ]
Reading and modifying current->mm and current->active_mm and switching mm should be done with irqs off, to prevent races seeing an intermediate state.
This is similar to commit 38cf307c1f20 ("mm: fix kthread_use_mm() vs TLB invalidate"). At exec-time when the new mm is activated, the old one should usually be single-threaded and no longer used, unless something else is holding an mm_users reference (which may be possible).
Absent other mm_users, there is also a race with preemption and lazy tlb switching. Consider the kernel_execve case where the current thread is using a lazy tlb active mm:
call_usermodehelper()
kernel_execve()
old_mm = current->mm;
active_mm = current->active_mm;
*** preempt *** --------------------> schedule()
prev->active_mm = NULL;
mmdrop(prev active_mm);
...
<-------------------- schedule()
current->mm = mm;
current->active_mm = mm;
if (!old_mm)
mmdrop(active_mm);If we switch back to the kernel thread from a different mm, there is a double free of the old active_mm, and a missing free of the new one.
Closing this race only requires interrupts to be disabled while ->mm and ->active_mm are being switched, but the TLB problem requires also holding interrupts off over activate_mm. Unfortunately not all archs can do that yet, e.g., arm defers the switch if irqs are disabled and expects finish_arch_post_lock_switch() to be called to complete the flush; um takes a blocking lock in activate_mm().
So as a first step, disable interrupts across the mm/active_mm updates to close the lazy tlb preempt race, and provide an arch option to extend that to activate_mm which allows architectures doing IPI based TLB shootdowns to close the second race.
This is a bit ugly, but in the interest of fixing the bug and backporting before all architectures are converted this is a compromise.
--- arch/Kconfig | 7 +++++++ fs/exec.c | 17 +++++++++++++++-- 2 files changed, 22 insertions(+), 2 deletions(-)
arch/Kconfig +7 -0
@@ -414,6 +414,13 @@ config MMU_GATHER_NO_GATHER bool depends on MMU_GATHER_TABLE_FREE +config ARCH_WANT_IRQS_OFF_ACTIVATE_MM + bool + help + Temporary select until all architectures can be converted to have + irqs disabled over activate_mm. Architectures that do IPI based TLB + shootdowns should enable this. + config ARCH_HAVE_NMI_SAFE_CMPXCHG bool
fs/exec.c +15 -2
@@ -1130,11 +1130,24 @@ static int exec_mmap(struct mm_struct *mm) }task_lock (tsk); - active_mm = tsk->active_mm;membarrier_exec_mmap (mm); - tsk->mm = mm; + +local_irq_disable (); + active_mm = tsk->active_mm; tsk->active_mm = mm; + tsk->mm = mm; +/* +* This prevents preemption while active_mm is being loaded and +* it and mm are being updated, which could cause problems for +* lazy tlb mm refcounting when these are updated by context +* switches. Not all architectures can handle irqs off over +* activate_mm yet. +*/ +if (!IS_ENABLED (CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM )) +local_irq_enable ();activate_mm (active_mm, mm); +if (IS_ENABLED (CONFIG_ARCH_WANT_IRQS_OFF_ACTIVATE_MM )) +local_irq_enable (); tsk->mm->vmacache_seqnum =0 ;vmacache_flush (tsk);task_unlock (tsk);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 7/147 powerpc: select ARCH_WANT_IRQS_OFF_ACTIVATE_MM
In reply to Sasha LevinFrom: Nicholas Piggin <[email protected]>
[ Upstream commit 66acd46080bd9e5ad2be4b0eb1d498d5145d058e ]
powerpc uses IPIs in some situations to switch a kernel thread away from a lazy tlb mm, which is subject to the TLB flushing race described in the changelog introducing ARCH_WANT_IRQS_OFF_ACTIVATE_MM.
--- arch/powerpc/Kconfig | 1 + arch/powerpc/include/asm/mmu_context.h | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-)
arch/powerpc/Kconfig +1 -0
@@ -148,6 +148,7 @@ config PPC select ARCH_USE_QUEUED_RWLOCKS if PPC_QUEUED_SPINLOCKS select ARCH_USE_QUEUED_SPINLOCKS if PPC_QUEUED_SPINLOCKS select ARCH_WANT_IPC_PARSE_VERSION + select ARCH_WANT_IRQS_OFF_ACTIVATE_MM select ARCH_WEAK_RELEASE_ACQUIRE select BINFMT_ELF select BUILDTIME_TABLE_SORT
arch/powerpc/include/asm/mmu_context.h +1 -1
@@ -244,7 +244,7 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,*/ static inline void activate_mm (struct mm_struct *prev,struct mm_struct *next) { -switch_mm (prev, next, current); +switch_mm_irqs_off (prev, next, current); }/* We don't currently use enter_lazy_tlb() for anything */
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 8/147 sparc64: remove mm_cpumask clearing to fix kthread_use_mm race
In reply to Sasha LevinFrom: Nicholas Piggin <[email protected]>
[ Upstream commit bafb056ce27940c9994ea905336aa8f27b4f7275 ]
The de facto (and apparently uncommented) standard for using an mm had, thanks to this code in sparc if nothing else, been that you must have a reference on mm_users *and that reference must have been obtained with mmget()*, i.e., from a thread with a reference to mm_users that had used the mm.
The introduction of mmget_not_zero() in commit d2005e3f41d4 ("userfaultfd: don't pin the user memory in userfaultfd_file_create()") allowed mm_count holders to aoperate on user mappings asynchronously from the actual threads using the mm, but they were not to load those mappings into their TLB (i.e., walking vmas and page tables is okay, kthread_use_mm() is not).
io_uring 2b188cc1bb857 ("Add io_uring IO interface") added code which does a kthread_use_mm() from a mmget_not_zero() refcount.
The problem with this is code which previously assumed mm = current->mm and mm->mm_users = 1 implies the mm will remain single-threaded at least until this thread creates another mm_users reference, has now broken.
arch/sparc/kernel/smp_64.c:
if (atomic_read(&mm->mm_users) = 1) {
cpumask_copy(mm_cpumask(mm), cpumask_of(cpu));
goto local_flush_and_out;
}vs fs/io_uring.c
if (unlikely(!(ctx->flags & IORING_SETUP_SQPOLL) ||
!mmget_not_zero(ctx->sqo_mm)))
return -EFAULT;
kthread_use_mm(ctx->sqo_mm);mmget_not_zero() could come in right after the mm_users = 1 test, then kthread_use_mm() which sets its CPU in the mm_cpumask. That update could be lost if cpumask_copy() occurs afterward.
I propose we fix this by allowing mmget_not_zero() to be a first-class reference, and not have this obscure undocumented and unchecked restriction.
The basic fix for sparc64 is to remove its mm_cpumask clearing code. The optimisation could be effectively restored by sending IPIs to mm_cpumask members and having them remove themselves from mm_cpumask. This is more tricky so I leave it as an exercise for someone with a sparc64 SMP. powerpc has a (currently similarly broken) example.
--- arch/sparc/kernel/smp_64.c | 65 ++++++++------------------------------ 1 file changed, 14 insertions(+), 51 deletions(-)
arch/sparc/kernel/smp_64.c +14 -51
@@ -1039,38 +1039,9 @@ void smp_fetch_global_pmu(void)* are flush_tlb_*() routines, and these run after flush_cache_*() * which performs the flushw. * -* The SMP TLB coherency scheme we use works as follows: -* -* 1) mm->cpu_vm_mask is a bit mask of which cpus an address -* space has (potentially) executed on, this is the heuristic -* we use to avoid doing cross calls. -* -* Also, for flushing from kswapd and also for clones, we -* use cpu_vm_mask as the list of cpus to make run the TLB. -* -* 2) TLB context numbers are shared globally across all processors -* in the system, this allows us to play several games to avoid -* cross calls. -* -* One invariant is that when a cpu switches to a process, and -* that processes tsk->active_mm->cpu_vm_mask does not have the -* current cpu's bit set, that tlb context is flushed locally. -* -* If the address space is non-shared (ie. mm->count = 1) we avoid -* cross calls when we want to flush the currently running process's -* tlb state. This is done by clearing all cpu bits except the current -* processor's in current->mm->cpu_vm_mask and performing the -* flush locally only. This will force any subsequent cpus which run -* this task to flush the context from the local tlb if the process -* migrates to another cpu (again). -* -* 3) For shared address spaces (threads) and swapping we bite the -* bullet for most cases and perform the cross call (but only to -* the cpus listed in cpu_vm_mask). -* -* The performance gain from "optimizing" away the cross call for threads is -* questionable (in theory the big win for threads is the massive sharing of -* address space state across processors). +* mm->cpu_vm_mask is a bit mask of which cpus an address +* space has (potentially) executed on, this is the heuristic +* we use to limit cross calls. */ /* This currently is only used by the hugetlb arch pre-fault
@@ -1080,18 +1051,13 @@ void smp_fetch_global_pmu(void)void smp_flush_tlb_mm (struct mm_struct *mm) {u32 ctx =CTX_HWBITS (mm->context); -int cpu =get_cpu (); -if (atomic_read (&mm->mm_users) =1 ) { -cpumask_copy (mm_cpumask (mm),cpumask_of (cpu)); - goto local_flush_and_out; - } +get_cpu ();smp_cross_call_masked (&xcall_flush_tlb_mm, ctx,0 ,0 ,mm_cpumask (mm)); -local_flush_and_out:__flush_tlb_mm (ctx,SECONDARY_CONTEXT );put_cpu ();
@@ -1114,17 +1080,15 @@ void smp_flush_tlb_pending(struct mm_struct *mm, unsigned long nr, unsigned long {u32 ctx =CTX_HWBITS (mm->context);struct tlb_pending_info info; -int cpu =get_cpu (); + +get_cpu (); info.ctx = ctx; info.nr = nr; info.vaddrs = vaddrs; -if (mm = current->mm &&atomic_read (&mm->mm_users) =1 ) -cpumask_copy (mm_cpumask (mm),cpumask_of (cpu)); -else -smp_call_function_many (mm_cpumask (mm), tlb_pending_func, - &info,1 ); +smp_call_function_many (mm_cpumask (mm), tlb_pending_func, + &info,1 );__flush_tlb_pending (ctx, nr, vaddrs);
@@ -1134,14 +1098,13 @@ void smp_flush_tlb_pending(struct mm_struct *mm, unsigned long nr, unsigned longvoid smp_flush_tlb_page (struct mm_struct *mm,unsigned long vaddr) {unsigned long context =CTX_HWBITS (mm->context); -int cpu =get_cpu (); -if (mm = current->mm &&atomic_read (&mm->mm_users) =1 ) -cpumask_copy (mm_cpumask (mm),cpumask_of (cpu)); -else -smp_cross_call_masked (&xcall_flush_tlb_page, - context, vaddr,0 , -mm_cpumask (mm)); +get_cpu (); + +smp_cross_call_masked (&xcall_flush_tlb_page, + context, vaddr,0 , +mm_cpumask (mm)); +__flush_tlb_page (context, vaddr);put_cpu ();
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 9/147 f2fs: add trace exit in exception path
In reply to Sasha LevinFrom: Zhang Qilong <[email protected]>
[ Upstream commit 9b66482282888d02832b7d90239e1cdb18e4b431 ]
Missing the trace exit in f2fs_sync_dirty_inodes
--- fs/f2fs/checkpoint.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
fs/f2fs/checkpoint.c +5 -1
@@ -1047,8 +1047,12 @@ int f2fs_sync_dirty_inodes(struct f2fs_sb_info *sbi, enum inode_type type)get_pages (sbi, is_dir ? F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA)); retry: -if (unlikely (f2fs_cp_error (sbi))) +if (unlikely (f2fs_cp_error (sbi))) { +trace_f2fs_sync_dirty_inodes_exit (sbi->sb, is_dir, +get_pages (sbi, is_dir ? +F2FS_DIRTY_DENTS :F2FS_DIRTY_DATA ));return -EIO ; + }spin_lock (&sbi->inode_lock[type]);
Signature · 8 lines
2.25.1 _______________________________________________ Linux-f2fs-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Sasha Levin
maintainerPATCH 10/147 f2fs: do sanity check on zoned block device path
In reply to Sasha LevinFrom: Chao Yu <[email protected]>
[ Upstream commit 07eb1d699452de04e9d389ff17fb8fc9e975d7bf ]
sbi->devs would be initialized only if image enables multiple device feature or blkzoned feature, if blkzoned feature flag was set by fuzz in non-blkzoned device, we will suffer below panic:
get_zone_idx fs/f2fs/segment.c:4892 [inline] f2fs_usable_zone_blks_in_seg fs/f2fs/segment.c:4943 [inline] f2fs_usable_blks_in_seg+0x39b/0xa00 fs/f2fs/segment.c:4999 Call Trace: check_block_count+0x69/0x4e0 fs/f2fs/segment.h:704 build_sit_entries fs/f2fs/segment.c:4403 [inline] f2fs_build_segment_manager+0x51da/0xa370 fs/f2fs/segment.c:5100 f2fs_fill_super+0x3880/0x6ff0 fs/f2fs/super.c:3684 mount_bdev+0x32e/0x3f0 fs/super.c:1417 legacy_get_tree+0x105/0x220 fs/fs_context.c:592 vfs_get_tree+0x89/0x2f0 fs/super.c:1547 do_new_mount fs/namespace.c:2896 [inline] path_mount+0x12ae/0x1e70 fs/namespace.c:3216 do_mount fs/namespace.c:3229 [inline] __do_sys_mount fs/namespace.c:3437 [inline] __se_sys_mount fs/namespace.c:3414 [inline] __x64_sys_mount+0x27f/0x300 fs/namespace.c:3414 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
Add sanity check to inconsistency on factors: blkzoned flag, device path and device character to avoid above panic.
--- fs/f2fs/super.c | 6 ++++++ 1 file changed, 6 insertions(+)
fs/f2fs/super.c +6 -0
@@ -2832,6 +2832,12 @@ static int sanity_check_raw_super(struct f2fs_sb_info *sbi, segment_count, dev_seg_count);return -EFSCORRUPTED ; } + }else { +if (__F2FS_HAS_FEATURE (raw_super,F2FS_FEATURE_BLKZONED ) && + !bdev_is_zoned (sbi->sb->s_bdev)) { +f2fs_info (sbi,"Zoned block device path is missing" ); +return -EFSCORRUPTED ; + } }if (secs_per_zone > total_sections || !secs_per_zone) {
Signature · 8 lines
2.25.1 _______________________________________________ Linux-f2fs-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Sasha Levin
maintainerPATCH 11/147 f2fs: fix uninit-value in f2fs_lookup
In reply to Sasha LevinFrom: Chao Yu <[email protected]>
[ Upstream commit 6d7ab88a98c1b7a47c228f8ffb4f44d631eaf284 ]
As syzbot reported:
Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x21c/0x280 lib/dump_stack.c:118 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:122 __msan_warning+0x58/0xa0 mm/kmsan/kmsan_instr.c:219 f2fs_lookup+0xe05/0x1a80 fs/f2fs/namei.c:503 lookup_open fs/namei.c:3082 [inline] open_last_lookups fs/namei.c:3177 [inline] path_openat+0x2729/0x6a90 fs/namei.c:3365 do_filp_open+0x2b8/0x710 fs/namei.c:3395 do_sys_openat2+0xa88/0x1140 fs/open.c:1168 do_sys_open fs/open.c:1184 [inline] __do_compat_sys_openat fs/open.c:1242 [inline] __se_compat_sys_openat+0x2a4/0x310 fs/open.c:1240 __ia32_compat_sys_openat+0x56/0x70 fs/open.c:1240 do_syscall_32_irqs_on arch/x86/entry/common.c:80 [inline] __do_fast_syscall_32+0x129/0x180 arch/x86/entry/common.c:139 do_fast_syscall_32+0x6a/0xc0 arch/x86/entry/common.c:162 do_SYSENTER_32+0x73/0x90 arch/x86/entry/common.c:205 entry_SYSENTER_compat_after_hwframe+0x4d/0x5c
In f2fs_lookup(), @res_page could be used before being initialized, because in __f2fs_find_entry(), once F2FS_I(dir)->i_current_depth was been fuzzed to zero, then @res_page will never be initialized, causing this kmsan warning, relocating @res_page initialization place to fix this bug.
--- fs/f2fs/dir.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-)
fs/f2fs/dir.c +3 -5
@@ -357,16 +357,15 @@ struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir,unsigned int max_depth;unsigned int level; + *res_page =NULL ; +if (f2fs_has_inline_dentry (dir)) { - *res_page =NULL ; de =f2fs_find_in_inline_dir (dir, fname, res_page); goto out; } -if (npages ==0 ) { - *res_page =NULL ; +if (npages ==0 ) goto out; - } max_depth =F2FS_I (dir)->i_current_depth;if (unlikely (max_depth >MAX_DIR_HASH_DEPTH )) {
@@ -377,7 +376,6 @@ struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir, }for (level =0 ; level < max_depth; level++) { - *res_page =NULL ; de =find_in_level (dir, level, fname, res_page);if (de ||IS_ERR (*res_page))break ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 12/147 f2fs: fix to check segment boundary during SIT page readahead
In reply to Sasha LevinFrom: Chao Yu <[email protected]>
[ Upstream commit 6a257471fa42c8c9c04a875cd3a2a22db148e0f0 ]
As syzbot reported:
kernel BUG at fs/f2fs/segment.h:657!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN
CPU: 1 PID: 16220 Comm: syz-executor.0 Not tainted 5.9.0-rc5-syzkaller #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:f2fs_ra_meta_pages+0xa51/0xdc0 fs/f2fs/segment.h:657 Call Trace: build_sit_entries fs/f2fs/segment.c:4195 [inline] f2fs_build_segment_manager+0x4b8a/0xa3c0 fs/f2fs/segment.c:4779 f2fs_fill_super+0x377d/0x6b80 fs/f2fs/super.c:3633 mount_bdev+0x32e/0x3f0 fs/super.c:1417 legacy_get_tree+0x105/0x220 fs/fs_context.c:592 vfs_get_tree+0x89/0x2f0 fs/super.c:1547 do_new_mount fs/namespace.c:2875 [inline] path_mount+0x1387/0x2070 fs/namespace.c:3192 do_mount fs/namespace.c:3205 [inline] __do_sys_mount fs/namespace.c:3413 [inline] __se_sys_mount fs/namespace.c:3390 [inline] __x64_sys_mount+0x27f/0x300 fs/namespace.c:3390 do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x44/0xa9
@blkno in f2fs_ra_meta_pages could exceed max segment count, causing panic in following sanity check in current_sit_addr(), add check condition to avoid this issue.
--- fs/f2fs/checkpoint.c | 2 ++ 1 file changed, 2 insertions(+)
fs/f2fs/checkpoint.c +2 -0
@@ -243,6 +243,8 @@ int f2fs_ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, blkno * NAT_ENTRY_PER_BLOCK);break ;case META_SIT : +if (unlikely (blkno >=TOTAL_SEGS (sbi))) + goto out;/* get sit block addr */ fio.new_blkaddr =current_sit_addr (sbi, blkno *SIT_ENTRY_PER_BLOCK );
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 13/147 s390/startup: avoid save_area_sync overflow
In reply to Sasha LevinFrom: Vasily Gorbik <[email protected]>
[ Upstream commit 2835c2ea95d50625108e47a459e1a47f6be836ce ]
Currently we overflow save_area_sync and write over save_area_async. Although this is not a real problem make startup_pgm_check_handler consistent with late pgm check handler and store [%r0,%r7] directly into gpregs_save_area.
--- arch/s390/boot/head.S | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-)
arch/s390/boot/head.S +11 -10
@@ -360,22 +360,23 @@ ENTRY(startup_kdump) # the save area and does disabled wait with a faulty address. # ENTRY(startup_pgm_check_handler) - stmg %r0,%r15,__LC_SAVE_AREA_SYNC - la %r1,4095 - stctg %c0,%c15,__LC_CREGS_SAVE_AREA-4095(%r1) - mvc __LC_GPREGS_SAVE_AREA-4095(128,%r1),__LC_SAVE_AREA_SYNC - mvc __LC_PSW_SAVE_AREA-4095(16,%r1),__LC_PGM_OLD_PSW + stmg %r8,%r15,__LC_SAVE_AREA_SYNC + la %r8,4095 + stctg %c0,%c15,__LC_CREGS_SAVE_AREA-4095(%r8) + stmg %r0,%r7,__LC_GPREGS_SAVE_AREA-4095(%r8) + mvc __LC_GPREGS_SAVE_AREA-4095+64(64,%r8),__LC_SAVE_AREA_SYNC + mvc __LC_PSW_SAVE_AREA-4095(16,%r8),__LC_PGM_OLD_PSW mvc __LC_RETURN_PSW(16),__LC_PGM_OLD_PSW ni __LC_RETURN_PSW,0xfc # remove IO and EX bits ni __LC_RETURN_PSW+1,0xfb # remove MCHK bit oi __LC_RETURN_PSW+1,0x2 # set wait state bit - larl %r2,.Lold_psw_disabled_wait - stg %r2,__LC_PGM_NEW_PSW+8 - l %r15,.Ldump_info_stack-.Lold_psw_disabled_wait(%r2) + larl %r9,.Lold_psw_disabled_wait + stg %r9,__LC_PGM_NEW_PSW+8 + l %r15,.Ldump_info_stack-.Lold_psw_disabled_wait(%r9) brasl %r14,print_pgm_check_info .Lold_psw_disabled_wait: - la %r1,4095 - lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r1) + la %r8,4095 + lmg %r0,%r15,__LC_GPREGS_SAVE_AREA-4095(%r8) lpswe __LC_RETURN_PSW # disabled wait .Ldump_info_stack: .long 0x5000 + PAGE_SIZE - STACK_FRAME_OVERHEAD
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 14/147 f2fs: compress: fix to disallow enabling compress on non-empty file
In reply to Sasha LevinFrom: Chao Yu <[email protected]>
[ Upstream commit 519a5a2f37b850f4eb86674a10d143088670a390 ]
Compressed inode and normal inode has different layout, so we should disallow enabling compress on non-empty file to avoid race condition during inode .i_addr array parsing and updating.
--- fs/f2fs/file.c | 2 ++ 1 file changed, 2 insertions(+)
fs/f2fs/file.c +2 -0
@@ -1836,6 +1836,8 @@ static int f2fs_setflags_common(struct inode *inode, u32 iflags, u32 mask)if (iflags &F2FS_COMPR_FL ) {if (!f2fs_may_compress (inode))return -EINVAL ; +if (S_ISREG (inode->i_mode) && inode->i_size) +return -EINVAL ;set_compress_context (inode); }
Signature · 8 lines
2.25.1 _______________________________________________ Linux-f2fs-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Sasha Levin
maintainerPATCH 15/147 s390/ap/zcrypt: revisit ap and zcrypt error handling
In reply to Sasha LevinFrom: Harald Freudenberger <[email protected]>
[ Upstream commit e0332629e33d1926c93348d918aaaf451ef9a16b ]
Revisit the ap queue error handling: Based on discussions and evaluatios with the firmware folk here is now a rework of the response code handling for all the AP instructions. The idea is to distinguish between failures because of some kind of invalid request where a retry does not make any sense and a failure where another attempt to send the very same request may succeed. The first case is handled by returning EINVAL to the userspace application. The second case results in retries within the zcrypt API controlled by a per message retry counter.
Revisit the zcrpyt error handling: Similar here, based on discussions with the firmware people here comes a rework of the handling of all the reply codes. Main point here is that there are only very few cases left, where a zcrypt device queue is switched to offline. It should never be the case that an AP reply message is 'unknown' to the device driver as it indicates a total mismatch between device driver and crypto card firmware. In all other cases, the code distinguishes between failure because of invalid message (see above - EINVAL) or failures of the infrastructure (see above - EAGAIN).
--- drivers/s390/crypto/ap_bus.h | 1 + drivers/s390/crypto/ap_queue.c | 8 +-- drivers/s390/crypto/zcrypt_debug.h | 8 +++ drivers/s390/crypto/zcrypt_error.h | 88 +++++++++--------------- drivers/s390/crypto/zcrypt_msgtype50.c | 50 +++++++------- drivers/s390/crypto/zcrypt_msgtype6.c | 92 +++++++++++++------------- 6 files changed, 116 insertions(+), 131 deletions(-)
drivers/s390/crypto/ap_bus.h +1 -0
@@ -50,6 +50,7 @@ static inline int ap_test_bit(unsigned int *ptr, unsigned int nr)#define AP_RESPONSE_NO_FIRST_PART 0x13#define AP_RESPONSE_MESSAGE_TOO_BIG 0x15#define AP_RESPONSE_REQ_FAC_NOT_INST 0x16 +#define AP_RESPONSE_INVALID_DOMAIN 0x42/* * Known device types
drivers/s390/crypto/ap_queue.c +3 -5
@@ -237,6 +237,9 @@ static enum ap_sm_wait ap_sm_write(struct ap_queue *aq)case AP_RESPONSE_RESET_IN_PROGRESS : aq->sm_state =AP_SM_STATE_RESET_WAIT ;return AP_SM_WAIT_TIMEOUT ; +case AP_RESPONSE_INVALID_DOMAIN : +AP_DBF (DBF_WARN ,"AP_RESPONSE_INVALID_DOMAIN on NQAP\n" ); + fallthrough;case AP_RESPONSE_MESSAGE_TOO_BIG :case AP_RESPONSE_REQ_FAC_NOT_INST :list_del_init (&ap_msg->list);
@@ -278,11 +281,6 @@ static enum ap_sm_wait ap_sm_reset(struct ap_queue *aq) aq->sm_state =AP_SM_STATE_RESET_WAIT ; aq->interrupt =AP_INTR_DISABLED ;return AP_SM_WAIT_TIMEOUT ; -case AP_RESPONSE_BUSY : -return AP_SM_WAIT_TIMEOUT ; -case AP_RESPONSE_Q_NOT_AVAIL : -case AP_RESPONSE_DECONFIGURED : -case AP_RESPONSE_CHECKSTOPPED :default : aq->sm_state =AP_SM_STATE_BORKED ;return AP_SM_WAIT_NONE ;
drivers/s390/crypto/zcrypt_debug.h +8 -0
@@ -21,6 +21,14 @@#define ZCRYPT_DBF (...) \ debug_sprintf_event(zcrypt_dbf_info, ##__VA_ARGS__) +#define ZCRYPT_DBF_ERR (...) \ + debug_sprintf_event(zcrypt_dbf_info, DBF_ERR, ##__VA_ARGS__) +#define ZCRYPT_DBF_WARN (...) \ + debug_sprintf_event(zcrypt_dbf_info, DBF_WARN, ##__VA_ARGS__) +#define ZCRYPT_DBF_INFO (...) \ + debug_sprintf_event(zcrypt_dbf_info, DBF_INFO, ##__VA_ARGS__) +#define ZCRYPT_DBF_DBG (...) \ + debug_sprintf_event(zcrypt_dbf_info, DBF_DEBUG, ##__VA_ARGS__)extern debug_info_t *zcrypt_dbf_info;
drivers/s390/crypto/zcrypt_error.h +32 -56
124 lines not on this page · raw message
drivers/s390/crypto/zcrypt_msgtype50.c +26 -24
82 lines not on this page · raw message
drivers/s390/crypto/zcrypt_msgtype6.c +46 -46
131 lines not on this page · raw message
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 16/147 um: change sigio_spinlock to a mutex
In reply to Sasha LevinFrom: Johannes Berg <[email protected]>
[ Upstream commit f2d05059e15af3f70502074f4e3a504530af504a ]
Lockdep complains at boot:
=============================
[ BUG: Invalid wait context ]
5.7.0-05093-g46d91ecd597b #98 Not tainted
-----------------------------
swapper/1 is trying to lock:0000000060931b98 (&desc[i].request_mutex){+.+.}-{3:3}, at: __setup_irq+0x11d/0x623 other info that might help us debug this:
context-{4:4}
1 lock held by swapper/1:
#0: 000000006074fed8 (sigio_spinlock){+.+.}-{2:2}, at: sigio_lock+0x1a/0x1c
stack backtrace: CPU: 0 PID: 1 Comm: swapper Not tainted 5.7.0-05093-g46d91ecd597b #98 Stack: 7fa4fab0 6028dfd1 0000002a 6008bea5 7fa50700 7fa50040 7fa4fac0 6028e016 7fa4fb50 6007f6da 60959c18 00000000 Call Trace: [<60023a0e>] show_stack+0x13b/0x155 [<6028e016>] dump_stack+0x2a/0x2c [<6007f6da>] __lock_acquire+0x515/0x15f2 [<6007eb50>] lock_acquire+0x245/0x273 [<6050d9f1>] __mutex_lock+0xbd/0x325 [<6050dc76>] mutex_lock_nested+0x1d/0x1f [<6008e27e>] __setup_irq+0x11d/0x623 [<6008e8ed>] request_threaded_irq+0x169/0x1a6 [<60021eb0>] um_request_irq+0x1ee/0x24b [<600234ee>] write_sigio_irq+0x3b/0x76 [<600383ca>] sigio_broken+0x146/0x2e4 [<60020bd8>] do_one_initcall+0xde/0x281
Because we hold sigio_spinlock and then get into requesting an interrupt with a mutex.
Change the spinlock to a mutex to avoid that.
--- arch/um/kernel/sigio.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
arch/um/kernel/sigio.c +3 -3
@@ -35,14 +35,14 @@ int write_sigio_irq(int fd) }/* These are called from os-Linux/sigio.c to protect its pollfds arrays. */ -static DEFINE_SPINLOCK (sigio_spinlock); +static DEFINE_MUTEX (sigio_mutex);void sigio_lock (void ) { -spin_lock (&sigio_spinlock); +mutex_lock (&sigio_mutex); }void sigio_unlock (void ) { -spin_unlock (&sigio_spinlock); +mutex_unlock (&sigio_mutex); }
Signature · 7 lines
2.25.1 _______________________________________________ linux-um mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-um
Sasha Levin
maintainerPATCH 17/147 f2fs: handle errors of f2fs_get_meta_page_nofail
In reply to Sasha LevinFrom: Jaegeuk Kim <[email protected]>
[ Upstream commit 86f33603f8c51537265ff7ac0320638fd2cbdb1b ]
First problem is we hit BUG_ON() in f2fs_get_sum_page given EIO on f2fs_get_meta_page_nofail().
Quick fix was not to give any error with infinite loop, but syzbot caught a case where it goes to that loop from fuzzed image. In turned out we abused f2fs_get_meta_page_nofail() like in the below call stack.
- f2fs_fill_super
- f2fs_build_segment_manager
- build_sit_entries
- get_current_sit_page
INFO: task syz-executor178:6870 can't die for more than 143 seconds.
task:syz-executor178 state:R
stack:26960 pid: 6870 ppid: 6869 flags:0x00004006
Call Trace:
Showing all locks held in the system:
1 lock held by khungtaskd/1179:
#0: ffffffff8a554da0 (rcu_read_lock){....}-{1:2}, at: debug_show_all_locks+0x53/0x260 kernel/locking/lockdep.c:6242
1 lock held by systemd-journal/3920:
1 lock held by in:imklog/6769:
#0: ffff88809eebc130 (&f->f_pos_lock){+.+.}-{3:3}, at: __fdget_pos+0xe9/0x100 fs/file.c:930
1 lock held by syz-executor178/6870:
#0: ffff8880925120e0 (&type->s_umount_key#47/1){+.+.}-{3:3}, at: alloc_super+0x201/0xaf0 fs/super.c:229Actually, we didn't have to use _nofail in this case, since we could return error to mount(2) already with the error handler.
As a result, this patch tries to 1) remove _nofail callers as much as possible,
2) deal with error case in last remaining caller, f2fs_get_sum_page().
--- fs/f2fs/checkpoint.c | 2 +- fs/f2fs/f2fs.h | 2 +- fs/f2fs/node.c | 2 +- fs/f2fs/segment.c | 12 +++++++++--- 4 files changed, 12 insertions(+), 6 deletions(-)
fs/f2fs/checkpoint.c +1 -1
@@ -107,7 +107,7 @@ struct page *f2fs_get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index)return __get_meta_page (sbi, index, true); } -struct page *f2fs_get_meta_page_nofail (struct f2fs_sb_info *sbi,pgoff_t index) +struct page *f2fs_get_meta_page_retry (struct f2fs_sb_info *sbi,pgoff_t index) {struct page *page;int count =0 ;
fs/f2fs/f2fs.h +1 -1
@@ -3385,7 +3385,7 @@ enum rw_hint f2fs_io_type_to_rw_hint(struct f2fs_sb_info *sbi,void f2fs_stop_checkpoint (struct f2fs_sb_info *sbi,bool end_io);struct page *f2fs_grab_meta_page (struct f2fs_sb_info *sbi,pgoff_t index);struct page *f2fs_get_meta_page (struct f2fs_sb_info *sbi,pgoff_t index); -struct page *f2fs_get_meta_page_nofail (struct f2fs_sb_info *sbi,pgoff_t index); +struct page *f2fs_get_meta_page_retry (struct f2fs_sb_info *sbi,pgoff_t index);struct page *f2fs_get_tmp_page (struct f2fs_sb_info *sbi,pgoff_t index);bool f2fs_is_valid_blkaddr (struct f2fs_sb_info *sbi,block_t blkaddr,int type);
fs/f2fs/node.c +1 -1
@@ -109,7 +109,7 @@ static void clear_node_page_dirty(struct page *page)static struct page *get_current_nat_page (struct f2fs_sb_info *sbi,nid_t nid) { -return f2fs_get_meta_page_nofail (sbi,current_nat_addr (sbi, nid)); +return f2fs_get_meta_page (sbi,current_nat_addr (sbi, nid)); }static struct page *get_next_nat_page (struct f2fs_sb_info *sbi,nid_t nid)
fs/f2fs/segment.c +9 -3
@@ -2344,7 +2344,9 @@ int f2fs_npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra)*/ struct page *f2fs_get_sum_page (struct f2fs_sb_info *sbi,unsigned int segno) { -return f2fs_get_meta_page_nofail (sbi,GET_SUM_BLOCK (sbi, segno)); +if (unlikely (f2fs_cp_error (sbi))) +return ERR_PTR (-EIO ); +return f2fs_get_meta_page_retry (sbi,GET_SUM_BLOCK (sbi, segno)); } void f2fs_update_meta_page(struct f2fs_sb_info *sbi,
@@ -2616,7 +2618,11 @@ static void change_curseg(struct f2fs_sb_info *sbi, int type)__next_free_blkoff (sbi, curseg,0 ); sum_page =f2fs_get_sum_page (sbi, new_segno); -f2fs_bug_on (sbi,IS_ERR (sum_page)); +if (IS_ERR (sum_page)) { +/* GC won't be able to use stale summary pages by cp_error */ +memset (curseg->sum_blk,0 ,SUM_ENTRY_SIZE ); +return ; + } sum_node = (struct f2fs_summary_block *)page_address (sum_page);memcpy (curseg->sum_blk, sum_node,SUM_ENTRY_SIZE );f2fs_put_page (sum_page,1 );
@@ -3781,7 +3787,7 @@ int f2fs_lookup_journal_in_cursum(struct f2fs_journal *journal, int type,static struct page *get_current_sit_page (struct f2fs_sb_info *sbi,unsigned int segno) { -return f2fs_get_meta_page_nofail (sbi,current_sit_addr (sbi, segno)); +return f2fs_get_meta_page (sbi,current_sit_addr (sbi, segno)); } static struct page *get_next_sit_page(struct f2fs_sb_info *sbi,
Signature · 8 lines
2.25.1 _______________________________________________ Linux-f2fs-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Sasha Levin
maintainerPATCH 18/147 afs: Don't assert on unpurgeable server records
In reply to Sasha LevinFrom: David Howells <[email protected]>
[ Upstream commit 7530d3eb3dcf1a30750e8e7f1f88b782b96b72b8 ]
Don't give an assertion failure on unpurgeable afs_server records - which kills the thread - but rather emit a trace line when we are purging a record (which only happens during network namespace removal or rmmod) and print a notice of the problem.
--- fs/afs/server.c | 7 ++++++- include/trace/events/afs.h | 2 ++ 2 files changed, 8 insertions(+), 1 deletion(-)
fs/afs/server.c +6 -1
@@ -550,7 +550,12 @@ void afs_manage_servers(struct work_struct *work)_debug ("manage %pU %u" , &server->uuid, active); - ASSERTIFCMP(purging, active, ==, 0); +if (purging) { +trace_afs_server (server,atomic_read (&server->ref), + active, afs_server_trace_purging); +if (active !=0 ) +pr_notice ("Can't purge s=%08x\n" , server->debug_id); + }if (active ==0 ) {time64_t expire_at = server->unuse_time;
include/trace/events/afs.h +2 -0
@@ -40,6 +40,7 @@ enum afs_server_trace { afs_server_trace_get_new_cbi, afs_server_trace_get_probe, afs_server_trace_give_up_cb, + afs_server_trace_purging, afs_server_trace_put_call, afs_server_trace_put_cbi, afs_server_trace_put_find_rsq,
@@ -270,6 +271,7 @@ enum afs_cb_break_reason { EM(afs_server_trace_get_new_cbi, "GET cbi ") \ EM(afs_server_trace_get_probe, "GET probe") \ EM(afs_server_trace_give_up_cb, "giveup-cb") \ + EM(afs_server_trace_purging, "PURGE ") \ EM(afs_server_trace_put_call, "PUT call ") \ EM(afs_server_trace_put_cbi, "PUT cbi ") \EM (afs_server_trace_put_find_rsq,"PUT f-rsq" ) \
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 19/147 MIPS: ftrace: Remove redundant #ifdef CONFIG_DYNAMIC_FTRACE
In reply to Sasha LevinFrom: Zejiang Tang <[email protected]>
[ Upstream commit 39116103a7345927fa99644d08bc0cc9d45fea6f ]
There exists redundant #ifdef CONFIG_DYNAMIC_FTRACE in ftrace.c, remove it.
--- arch/mips/kernel/ftrace.c | 4 ---- 1 file changed, 4 deletions(-)
arch/mips/kernel/ftrace.c +0 -4
@@ -37,10 +37,6 @@ void arch_ftrace_update_code(int command)ftrace_modify_all_code (command); } -#endif - -#ifdef CONFIG_DYNAMIC_FTRACE -#define JAL 0x0c000000/* jump & link: ip --> ra, jump to target */ #define ADDR_MASK 0x03ffffff/* op_code|addr : 31...26|25 ....0 */ #define JUMP_RANGE_MASK ((1UL << 28) - 1)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 20/147 powerpc/64s: handle ISA v3.1 local copy-paste context switches
In reply to Sasha LevinFrom: Nicholas Piggin <[email protected]>
[ Upstream commit dc462267d2d7aacffc3c1d99b02d7a7c59db7c66 ]
The ISA v3.1 the copy-paste facility has a new memory move functionality which allows the copy buffer to be pasted to domestic memory (RAM) as opposed to foreign memory (accelerator).
This means the POWER9 trick of avoiding the cp_abort on context switch if the process had not mapped foreign memory does not work on POWER10. Do the cp_abort unconditionally there.
KVM must also cp_abort on guest exit to prevent copy buffer state leaking between contexts.
--- arch/powerpc/kernel/process.c | 16 +++++++++------- arch/powerpc/kvm/book3s_hv.c | 7 +++++++ arch/powerpc/kvm/book3s_hv_rmhandlers.S | 8 ++++++++ 3 files changed, 24 insertions(+), 7 deletions(-)
arch/powerpc/kernel/process.c +9 -7
@@ -1256,15 +1256,17 @@ struct task_struct *__switch_to(struct task_struct *prev,restore_math (current->thread.regs);/* -* The copy-paste buffer can only store into foreign real -* addresses, so unprivileged processes can not see the -* data or use it in any way unless they have foreign real -* mappings. If the new process has the foreign real address -* mappings, we must issue a cp_abort to clear any state and -* prevent snooping, corruption or a covert channel. +* On POWER9 the copy-paste buffer can only paste into +* foreign real addresses, so unprivileged processes can not +* see the data or use it in any way unless they have +* foreign real mappings. If the new process has the foreign +* real address mappings, we must issue a cp_abort to clear +* any state and prevent snooping, corruption or a covert +* channel. ISA v3.1 supports paste into local memory. */ if (current->mm && - atomic_read(¤t->mm->context.vas_windows)) + (cpu_has_feature (CPU_FTR_ARCH_31 ) || +atomic_read (¤t->mm->context.vas_windows))) asm volatile(PPC_CP_ABORT); } #endif /* CONFIG_PPC_BOOK3S_64 */
arch/powerpc/kvm/book3s_hv.c +7 -0
@@ -3530,6 +3530,13 @@ static int kvmhv_load_hv_regs_and_go(struct kvm_vcpu *vcpu, u64 time_limit,*/ asmvolatile ("eieio; tlbsync; ptesync" ); +/* +* cp_abort is required if the processor supports local copy-paste +* to clear the copy buffer that was under control of the guest. +*/ +if (cpu_has_feature (CPU_FTR_ARCH_31 )) + asm volatile(PPC_CP_ABORT); +mtspr (SPRN_LPID , vcpu->kvm->arch.host_lpid);/* restore host LPID */ isync ();
arch/powerpc/kvm/book3s_hv_rmhandlers.S +8 -0
@@ -1830,6 +1830,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_P9_RADIX_PREFETCH_BUG) 2: #endif /* CONFIG_PPC_RADIX_MMU */ + /* + * cp_abort is required if the processor supports local copy-paste + * to clear the copy buffer that was under control of the guest. + */ +BEGIN_FTR_SECTION + PPC_CP_ABORT +END_FTR_SECTION_IFSET(CPU_FTR_ARCH_31) + /* * POWER7/POWER8 guest -> host partition switch code. * We don't have to lock against tlbies but we do
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 21/147 ARM: 8997/2: hw_breakpoint: Handle inexact watchpoint addresses
In reply to Sasha LevinFrom: Douglas Anderson <[email protected]>
[ Upstream commit 22c9e58299e5f18274788ce54c03d4fb761e3c5d ]
This is commit fdfeff0f9e3d ("arm64: hw_breakpoint: Handle inexact watchpoint addresses") but ported to arm32, which has the same problem.
This problem was found by Android CTS tests, notably the "watchpoint_imprecise" test [1]. I tested locally against a copycat (simplified) version of the test though.
[1] https://android.googlesource.com/platform/bionic/+/master/tests/sys_ptrace_test.cpp
--- arch/arm/kernel/hw_breakpoint.c | 100 +++++++++++++++++++++++--------- 1 file changed, 72 insertions(+), 28 deletions(-)
arch/arm/kernel/hw_breakpoint.c +72 -28
@@ -683,6 +683,40 @@ static void disable_single_step(struct perf_event *bp)arch_install_hw_breakpoint (bp); } +/* +* Arm32 hardware does not always report a watchpoint hit address that matches +* one of the watchpoints set. It can also report an address "near" the +* watchpoint if a single instruction access both watched and unwatched +* addresses. There is no straight-forward way, short of disassembling the +* offending instruction, to map that address back to the watchpoint. This +* function computes the distance of the memory access from the watchpoint as a +* heuristic for the likelyhood that a given access triggered the watchpoint. +* +* See this same function in the arm64 platform code, which has the same +* problem. +* +* The function returns the distance of the address from the bytes watched by +* the watchpoint. In case of an exact match, it returns 0. +*/ +static u32 get_distance_from_watchpoint (unsigned long addr,u32 val, +struct arch_hw_breakpoint_ctrl *ctrl) +{ +u32 wp_low, wp_high; +u32 lens, lene; + + lens =__ffs (ctrl->len); + lene =__fls (ctrl->len); + + wp_low = val + lens; + wp_high = val + lene; +if (addr < wp_low) +return wp_low - addr; +else if (addr > wp_high) +return addr - wp_high; +else +return 0 ; +} +static int watchpoint_fault_on_uaccess (struct pt_regs *regs,struct arch_hw_breakpoint *info) {
@@ -692,23 +726,25 @@ static int watchpoint_fault_on_uaccess(struct pt_regs *regs,static void watchpoint_handler (unsigned long addr,unsigned int fsr,struct pt_regs *regs) { -int i, access; -u32 val, ctrl_reg, alignment_mask; +int i, access, closest_match =0 ; +u32 min_dist =-1 , dist; +u32 val, ctrl_reg;struct perf_event *wp, **slots;struct arch_hw_breakpoint *info;struct arch_hw_breakpoint_ctrl ctrl; slots =this_cpu_ptr (wp_on_reg); +/* +* Find all watchpoints that match the reported address. If no exact +* match is found. Attribute the hit to the closest watchpoint. +*/ +rcu_read_lock ();for (i =0 ; i < core_num_wrps; ++i) { -rcu_read_lock (); - wp = slots[i]; -if (wp ==NULL ) - goto unlock; +continue ; - info =counter_arch_bp (wp);/* * The DFAR is an unknown value on debug architectures prior * to 7.1. Since we only allow a single watchpoint on these
@@ -717,33 +753,31 @@ static void watchpoint_handler(unsigned long addr, unsigned int fsr,*/ if (debug_arch <ARM_DEBUG_ARCH_V7_1 ) {BUG_ON (i >0 ); + info =counter_arch_bp (wp); info->trigger = wp->attr.bp_addr; }else { -if (info->ctrl.len ==ARM_BREAKPOINT_LEN_8 ) - alignment_mask =0x7 ; -else - alignment_mask =0x3 ; - -/* Check if the watchpoint value matches. */ - val =read_wb_reg (ARM_BASE_WVR + i); -if (val != (addr & ~alignment_mask)) - goto unlock; - -/* Possible match, check the byte address select. */ - ctrl_reg =read_wb_reg (ARM_BASE_WCR + i); -decode_ctrl_reg (ctrl_reg, &ctrl); -if (!((1 << (addr & alignment_mask)) & ctrl.len)) - goto unlock; -/* Check that the access type matches. */ if (debug_exception_updates_fsr ()) { access = (fsr &ARM_FSR_ACCESS_MASK ) ?HW_BREAKPOINT_W :HW_BREAKPOINT_R ;if (!(access &hw_breakpoint_type (wp))) - goto unlock; +continue ; } + val =read_wb_reg (ARM_BASE_WVR + i); + ctrl_reg =read_wb_reg (ARM_BASE_WCR + i); +decode_ctrl_reg (ctrl_reg, &ctrl); + dist =get_distance_from_watchpoint (addr, val, &ctrl); +if (dist < min_dist) { + min_dist = dist; + closest_match = i; + } +/* Is this an exact match? */ +if (dist !=0 ) +continue ; +/* We have a winner. */ + info =counter_arch_bp (wp); info->trigger = addr; }
@@ -765,13 +799,23 @@ static void watchpoint_handler(unsigned long addr, unsigned int fsr,* we can single-step over the watchpoint trigger. */ if (!is_default_overflow_handler (wp)) - goto unlock; - +continue ; step:enable_single_step (wp,instruction_pointer (regs)); -unlock: -rcu_read_unlock (); } + +if (min_dist >0 && min_dist !=-1 ) { +/* No exact match found. */ + wp = slots[closest_match]; + info =counter_arch_bp (wp); + info->trigger = addr; +pr_debug ("watchpoint fired: address = 0x%x\n" , info->trigger); +perf_bp_event (wp, regs); +if (is_default_overflow_handler (wp)) +enable_single_step (wp,instruction_pointer (regs)); + } + +rcu_read_unlock (); }static void watchpoint_single_step_handler (unsigned long pc)
Signature · 7 lines
2.25.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Sasha Levin
maintainerPATCH 22/147 NFS4: Fix oops when copy_file_range is attempted with NFS4.0 source
In reply to Sasha LevinFrom: Dave Wysochanski <[email protected]>
[ Upstream commit d8a6ad913c286d4763ae20b14c02fe6f39d7cd9f ]
The following oops is seen during xfstest/565 when the 'test' (source of the copy) is NFS4.0 and 'scratch' (destination) is NFS4.2
[ 59.692458] run fstests generic/565 at 2020-08-01 05:50:35 [ 60.613588] BUG: kernel NULL pointer dereference, address: 0000000000000008 [ 60.624970] #PF: supervisor read access in kernel mode [ 60.627671] #PF: error_code(0x0000) - not-present page [ 60.630347] PGD 0 P4D 0 [ 60.631853] Oops: 0000 [#1] SMP PTI [ 60.634086] CPU: 6 PID: 2828 Comm: xfs_io Kdump: loaded Not tainted 5.8.0-rc3 #1 [ 60.637676] Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011 [ 60.639901] RIP: 0010:nfs4_check_serverowner_major_id+0x5/0x30 [nfsv4] [ 60.642719] Code: 89 ff e8 3e b3 b8 e1 e9 71 fe ff ff 41 bc da d8 ff ff e9 c3 fe ff ff e8 e9 9d 08 e2 66 0f 1f 84 00 00 00 00 00 66 66 66 66 90 <8b> 57 08 31 c0 3b 56 08 75 12 48 83 c6 0c 48 83 c7 0c e8 c4 97 bb [ 60.652629] RSP: 0018:ffffc265417f7e10 EFLAGS: 00010287 [ 60.655379] RAX: ffffa0664b066400 RBX: 0000000000000000 RCX: 0000000000000001 [ 60.658754] RDX: ffffa066725fb000 RSI: ffffa066725fd000 RDI: 0000000000000000 [ 60.662292] RBP: 0000000000020000 R08: 0000000000020000 R09: 0000000000000000 [ 60.666189] R10: 0000000000000003 R11: 0000000000000000 R12: ffffa06648258d00 [ 60.669914] R13: 0000000000000000 R14: 0000000000000000 R15: ffffa06648258100 [ 60.673645] FS: 00007faa9fb35800(0000) GS:ffffa06677d80000(0000) knlGS:0000000000000000 [ 60.677698] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 60.680773] CR2: 0000000000000008 CR3: 0000000203f14000 CR4: 00000000000406e0 [ 60.684476] Call Trace: [ 60.685809] nfs4_copy_file_range+0xfc/0x230 [nfsv4] [ 60.688704] vfs_copy_file_range+0x2ee/0x310 [ 60.691104] __x64_sys_copy_file_range+0xd6/0x210 [ 60.693527] do_syscall_64+0x4d/0x90 [ 60.695512] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 60.698006] RIP: 0033:0x7faa9febc1bd
--- fs/nfs/nfs4file.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
fs/nfs/nfs4file.c +2 -1
@@ -145,7 +145,8 @@ static ssize_t __nfs4_copy_file_range(struct file *file_in, loff_t pos_in,/* Only offload copy if superblock is the same */ if (file_in->f_op != &nfs4_file_operations)return -EXDEV ; -if (!nfs_server_capable (file_inode (file_out),NFS_CAP_COPY )) +if (!nfs_server_capable (file_inode (file_out),NFS_CAP_COPY ) || + !nfs_server_capable (file_inode (file_in),NFS_CAP_COPY ))return -EOPNOTSUPP ;if (file_inode (file_in) ==file_inode (file_out))return -EOPNOTSUPP ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 23/147 xfs: Set xfs_buf type flag when growing summary/bitmap files
In reply to Sasha LevinFrom: Chandan Babu R <[email protected]>
[ Upstream commit 72cc95132a93293dcd0b6f68353f4741591c9aeb ]
The following sequence of commands,
mkfs.xfs -f -m reflink=0 -r rtdev=/dev/loop1,size=10M /dev/loop0 mount -o rtdev=/dev/loop1 /dev/loop0 /mnt
xfs_growfs /mnt
... causes the following call trace to be printed on the console,
XFS: Assertion failed: (bip->bli_flags & XFS_BLI_STALE) || (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF), file: fs/xfs/xfs_buf_item.c, line: 331
Call Trace: xfs_buf_item_format+0x632/0x680 ? kmem_alloc_large+0x29/0x90 ? kmem_alloc+0x70/0x120 ? xfs_log_commit_cil+0x132/0x940 xfs_log_commit_cil+0x26f/0x940 ? xfs_buf_item_init+0x1ad/0x240 ? xfs_growfs_rt_alloc+0x1fc/0x280 __xfs_trans_commit+0xac/0x370 xfs_growfs_rt_alloc+0x1fc/0x280 xfs_growfs_rt+0x1a0/0x5e0 xfs_file_ioctl+0x3fd/0xc70 ? selinux_file_ioctl+0x174/0x220 ksys_ioctl+0x87/0xc0 __x64_sys_ioctl+0x16/0x20 do_syscall_64+0x3e/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xa9
This occurs because the buffer being formatted has the value of XFS_BLFT_UNKNOWN_BUF assigned to the 'type' subfield of bip->bli_formats->blf_flags.
This commit fixes the issue by assigning one of XFS_BLFT_RTSUMMARY_BUF and XFS_BLFT_RTBITMAP_BUF to the 'type' subfield of bip->bli_formats->blf_flags before committing the corresponding transaction.
--- fs/xfs/xfs_rtalloc.c | 8 ++++++++ 1 file changed, 8 insertions(+)
fs/xfs/xfs_rtalloc.c +8 -0
@@ -767,8 +767,14 @@ xfs_growfs_rt_alloc(struct xfs_bmbt_irec map;/* block map output */ int nmap;/* number of block maps */ int resblks;/* space reservation */ +enum xfs_blft buf_type;struct xfs_trans *tp; +if (ip == mp->m_rsumip) + buf_type =XFS_BLFT_RTSUMMARY_BUF ; +else + buf_type =XFS_BLFT_RTBITMAP_BUF ; +/* * Allocate space to the file, as necessary. */
@@ -830,6 +836,8 @@ xfs_growfs_rt_alloc( mp->m_bsize, 0, &bp);if (error) goto out_trans_cancel; + +xfs_trans_buf_set_type (tp, bp, buf_type);memset (bp->b_addr,0 , mp->m_sb.sb_blocksize);xfs_trans_log_buf (tp, bp,0 , mp->m_sb.sb_blocksize -1 );/*
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 24/147 xfs: Set xfs_buf's b_ops member when zeroing bitmap/summary files
In reply to Sasha LevinFrom: Chandan Babu R <[email protected]>
[ Upstream commit c54e14d155f5fdbac73a8cd4bd2678cb252149dc ]
In xfs_growfs_rt(), we enlarge bitmap and summary files by allocating new blocks for both files. For each of the new blocks allocated, we allocate an xfs_buf, zero the payload, log the contents and commit the transaction. Hence these buffers will eventually find themselves appended to list at xfs_ail->ail_buf_list.
Later, xfs_growfs_rt() loops across all of the new blocks belonging to the bitmap inode to set the bitmap values to 1. In doing so, it allocates a new transaction and invokes the following sequence of functions,
- xfs_rtfree_range()
- xfs_rtmodify_range()
- xfs_rtbuf_get()
We pass '&xfs_rtbuf_ops' as the ops pointer to xfs_trans_read_buf().
- xfs_trans_read_buf()
We find the xfs_buf of interest in per-ag hash table, invoke xfs_buf_reverify() which ends up assigning '&xfs_rtbuf_ops' to xfs_buf->b_ops.
On the other hand, if xfs_growfs_rt_alloc() had allocated a few blocks for the bitmap inode and returned with an error, all the xfs_bufs corresponding to the new bitmap blocks that have been allocated would continue to be on xfs_ail->ail_buf_list list without ever having a non-NULL value assigned to their b_ops members. An AIL flush operation would then trigger the following warning message to be printed on the console,
XFS (loop0): _xfs_buf_ioapply: no buf ops on daddr 0x58 len 8
00000000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00000020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00000030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00000040: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00000050: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00000060: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00000070: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ CPU: 3 PID: 449 Comm: xfsaild/loop0 Not tainted 5.8.0-rc4-chandan-00038-g4d8c2b9de9ab-dirty #37 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 Call Trace: dump_stack+0x57/0x70 _xfs_buf_ioapply+0x37c/0x3b0 ? xfs_rw_bdev+0x1e0/0x1e0 ? xfs_buf_delwri_submit_buffers+0xd4/0x210 __xfs_buf_submit+0x6d/0x1f0 xfs_buf_delwri_submit_buffers+0xd4/0x210 xfsaild+0x2c8/0x9e0 ? __switch_to_asm+0x42/0x70 ? xfs_trans_ail_cursor_first+0x80/0x80 kthread+0xfe/0x140 ? kthread_park+0x90/0x90 ret_from_fork+0x22/0x30
This message indicates that the xfs_buf had its b_ops member set to NULL.
This commit fixes the issue by assigning "&xfs_rtbuf_ops" to b_ops member of each of the xfs_bufs logged by xfs_growfs_rt_alloc().
--- fs/xfs/xfs_rtalloc.c | 1 + 1 file changed, 1 insertion(+)
fs/xfs/xfs_rtalloc.c +1 -0
@@ -838,6 +838,7 @@ xfs_growfs_rt_alloc( goto out_trans_cancel;xfs_trans_buf_set_type (tp, bp, buf_type); + bp->b_ops = &xfs_rtbuf_ops;memset (bp->b_addr,0 , mp->m_sb.sb_blocksize);xfs_trans_log_buf (tp, bp,0 , mp->m_sb.sb_blocksize -1 );/*
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 25/147 xfs: log new intent items created as part of finishing recovered intent items
In reply to Sasha LevinFrom: "Darrick J. Wong" <[email protected]>
[ Upstream commit 93293bcbde93567efaf4e6bcd58cad270e1fcbf5 ]
During a code inspection, I found a serious bug in the log intent item recovery code when an intent item cannot complete all the work and decides to requeue itself to get that done. When this happens, the item recovery creates a new incore deferred op representing the remaining work and attaches it to the transaction that it allocated. At the end of _item_recover, it moves the entire chain of deferred ops to the dummy parent_tp that xlog_recover_process_intents passed to it, but fail to log a new intent item for the remaining work before committing the transaction for the single unit of work.
xlog_finish_defer_ops logs those new intent items once recovery has finished dealing with the intent items that it recovered, but this isn't sufficient. If the log is forced to disk after a recovered log item decides to requeue itself and the system goes down before we call xlog_finish_defer_ops, the second log recovery will never see the new intent item and therefore has no idea that there was more work to do. It will finish recovery leaving the filesystem in a corrupted state.
The same logic applies to /any/ deferred ops added during intent item recovery, not just the one handling the remaining work.
--- fs/xfs/libxfs/xfs_defer.c | 26 ++++++++++++++++++++++++-- fs/xfs/libxfs/xfs_defer.h | 6 ++++++ fs/xfs/xfs_bmap_item.c | 2 +- fs/xfs/xfs_refcount_item.c | 2 +- 4 files changed, 32 insertions(+), 4 deletions(-)
fs/xfs/libxfs/xfs_defer.c +24 -2
@@ -186,8 +186,9 @@ xfs_defer_create_intent( {const struct xfs_defer_op_type *ops = defer_op_types[dfp->dfp_type]; - dfp->dfp_intent = ops->create_intent (tp, &dfp->dfp_work, - dfp->dfp_count, sort); +if (!dfp->dfp_intent) + dfp->dfp_intent = ops->create_intent (tp, &dfp->dfp_work, + dfp->dfp_count, sort); }/*
@@ -390,6 +391,7 @@ xfs_defer_finish_one(list_add (li, &dfp->dfp_work); dfp->dfp_count++; dfp->dfp_done =NULL ; + dfp->dfp_intent =NULL ;xfs_defer_create_intent (tp, dfp, false); }
@@ -552,3 +554,23 @@ xfs_defer_move(xfs_defer_reset (stp); } + +/* +* Prepare a chain of fresh deferred ops work items to be completed later. Log +* recovery requires the ability to put off until later the actual finishing +* work so that it can process unfinished items recovered from the log in +* correct order. +* +* Create and log intent items for all the work that we're capturing so that we +* can be assured that the items will get replayed if the system goes down +* before log recovery gets a chance to finish the work it put off. Then we +* move the chain from stp to dtp. +*/ +void +xfs_defer_capture ( +struct xfs_trans *dtp, +struct xfs_trans *stp) +{ +xfs_defer_create_intents (stp); +xfs_defer_move (dtp, stp); +}
fs/xfs/libxfs/xfs_defer.h +6 -0
@@ -63,4 +63,10 @@ extern const struct xfs_defer_op_type xfs_rmap_update_defer_type;extern const struct xfs_defer_op_type xfs_extent_free_defer_type;extern const struct xfs_defer_op_type xfs_agfl_free_defer_type; +/* +* Functions to capture a chain of deferred operations and continue them later. +* This doesn't normally happen except log recovery. +*/ +void xfs_defer_capture (struct xfs_trans *dtp,struct xfs_trans *stp); +#endif /* __XFS_DEFER_H__ */
fs/xfs/xfs_bmap_item.c +1 -1
@@ -534,7 +534,7 @@ xfs_bui_item_recover(xfs_bmap_unmap_extent (tp, ip, &irec); } -xfs_defer_move (parent_tp, tp); +xfs_defer_capture (parent_tp, tp); error =xfs_trans_commit (tp);xfs_iunlock (ip,XFS_ILOCK_EXCL );xfs_irele (ip);
fs/xfs/xfs_refcount_item.c +1 -1
@@ -555,7 +555,7 @@ xfs_cui_item_recover( }xfs_refcount_finish_one_cleanup (tp, rcur, error); -xfs_defer_move (parent_tp, tp); +xfs_defer_capture (parent_tp, tp); error =xfs_trans_commit (tp);return error;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 26/147 power: supply: bq27xxx: report "not charging" on all types
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 7bf738ba110722b63e9dc8af760d3fb2aef25593 ]
Commit 6f24ff97e323 ("power: supply: bq27xxx_battery: Add the BQ27Z561 Battery monitor") and commit d74534c27775 ("power:bq27xxx_battery: Add support for additional bq27xxx family devices") added support for new device types by copying most of the code and adding necessary quirks.
However they did not copy the code in bq27xxx_battery_status() responsible for returning POWER_SUPPLY_STATUS_NOT_CHARGING.
Unify the bq27xxx_battery_status() so for all types when charger is supplied, it will return "not charging" status.
--- drivers/power/supply/bq27xxx_battery.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/power/supply/bq27xxx_battery.c +4 -2
@@ -1772,8 +1772,6 @@ static int bq27xxx_battery_status(struct bq27xxx_device_info *di, status =POWER_SUPPLY_STATUS_FULL ; else if (di->cache.flags & BQ27000_FLAG_CHGS) status =POWER_SUPPLY_STATUS_CHARGING ; - else if (power_supply_am_i_supplied(di->bat) > 0) - status =POWER_SUPPLY_STATUS_NOT_CHARGING ; else status =POWER_SUPPLY_STATUS_DISCHARGING ; }else if (di->opts &BQ27Z561_O_BITS ) {
@@ -1792,6 +1790,10 @@ static int bq27xxx_battery_status(struct bq27xxx_device_info *di, status =POWER_SUPPLY_STATUS_CHARGING ; } +if ((status ==POWER_SUPPLY_STATUS_DISCHARGING ) && + (power_supply_am_i_supplied (di->bat) >0 )) + status =POWER_SUPPLY_STATUS_NOT_CHARGING ; + val->intval = status;return 0 ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 27/147 xfs: change the order in which child and parent defer ops are finished
In reply to Sasha LevinFrom: "Darrick J. Wong" <[email protected]>
[ Upstream commit 27dada070d59c28a441f1907d2cec891b17dcb26 ]
The defer ops code has been finishing items in the wrong order -- if a top level defer op creates items A and B, and finishing item A creates more defer ops A1 and A2, we'll put the new items on the end of the chain and process them in the order A B A1 A2. This is kind of weird, since it's convenient for programmers to be able to think of A and B as an ordered sequence where all the sub-tasks for A must finish before we move on to B, e.g. A A1 A2 D.
Right now, our log intent items are not so complex that this matters, but this will become important for the atomic extent swapping patchset. In order to maintain correct reference counting of extents, we have to unmap and remap extents in that order, and we want to complete that work before moving on to the next range that the user wants to swap. This patch fixes defer ops to satsify that requirement.
The primary symptom of the incorrect order was noticed in an early performance analysis of the atomic extent swap code. An astonishingly large number of deferred work items accumulated when userspace requested an atomic update of two very fragmented files. The cause of this was traced to the same ordering bug in the inner loop of xfs_defer_finish_noroll.
If the ->finish_item method of a deferred operation queues new deferred operations, those new deferred ops are appended to the tail of the pending work list. To illustrate, say that a caller creates a transaction t0 with four deferred operations D0-D3. The first thing defer ops does is roll the transaction to t1, leaving us with:
t1: D0(t0), D1(t0), D2(t0), D3(t0)
Let's say that finishing each of D0-D3 will create two new deferred ops. After finish D0 and roll, we'll have the following chain:
t2: D1(t0), D2(t0), D3(t0), d4(t1), d5(t1)
d4 and d5 were logged to t1. Notice that while we're about to start work on D1, we haven't actually completed all the work implied by D0 being finished. So far we've been careful (or lucky) to structure the dfops callers such that D1 doesn't depend on d4 or d5 being finished, but this is a potential logic bomb.
There's a second problem lurking. Let's see what happens as we finish D1-D3:
t3: D2(t0), D3(t0), d4(t1), d5(t1), d6(t2), d7(t2)
t4: D3(t0), d4(t1), d5(t1), d6(t2), d7(t2), d8(t3), d9(t3)
t5: d4(t1), d5(t1), d6(t2), d7(t2), d8(t3), d9(t3), d10(t4), d11(t4)
Let's say that d4-d11 are simple work items that don't queue any other operations, which means that we can complete each d4 and roll to t6:
t6: d5(t1), d6(t2), d7(t2), d8(t3), d9(t3), d10(t4), d11(t4)
t7: d6(t2), d7(t2), d8(t3), d9(t3), d10(t4), d11(t4)
...
t11: d10(t4), d11(t4)
t12: d11(t4)
<done>
When we try to roll to transaction #12, we're holding defer op d11, which we logged way back in t4. This means that the tail of the log is pinned at t4. If the log is very small or there are a lot of other threads updating metadata, this means that we might have wrapped the log and cannot get roll to t11 because there isn't enough space left before we'd run into t4.
Let's shift back to the original failure. I mentioned before that I discovered this flaw while developing the atomic file update code. In that scenario, we have a defer op (D0) that finds a range of file blocks to remap, creates a handful of new defer ops to do that, and then asks to be continued with however much work remains.
So, D0 is the original swapext deferred op. The first thing defer ops does is rolls to t1:
t1: D0(t0)
We try to finish D0, logging d1 and d2 in the process, but can't get all the work done. We log a done item and a new intent item for the work that D0 still has to do, and roll to t2:
t2: D0'(t1), d1(t1), d2(t1)
We roll and try to finish D0', but still can't get all the work done, so we log a done item and a new intent item for it, requeue D0 a second time, and roll to t3:
t3: D0''(t2), d1(t1), d2(t1), d3(t2), d4(t2)
If it takes 48 more rolls to complete D0, then we'll finally dispense with D0 in t50:
t50: D<fifty primes>(t49), d1(t1), ..., d102(t50)
We then try to roll again to get a chain like this:
t51: d1(t1), d2(t1), ..., d101(t50), d102(t50)
...
t152: d102(t50)
<done>
Notice that in rolling to transaction #51, we're holding on to a log intent item for d1 that was logged in transaction #1. This means that the tail of the log is pinned at t1. If the log is very small or there are a lot of other threads updating metadata, this means that we might have wrapped the log and cannot roll to t51 because there isn't enough space left before we'd run into t1. This is of course problem #2 again.
But notice the third problem with this scenario: we have 102 defer ops tied to this transaction! Each of these items are backed by pinned kernel memory, which means that we risk OOM if the chains get too long.
Yikes. Problem #1 is a subtle logic bomb that could hit someone in the future; problem #2 applies (rarely) to the current upstream, and problem #3 applies to work under development.
This is not how incremental deferred operations were supposed to work. The dfops design of logging in the same transaction an intent-done item and a new intent item for the work remaining was to make it so that we only have to juggle enough deferred work items to finish that one small piece of work. Deferred log item recovery will find that first unfinished work item and restart it, no matter how many other intent items might follow it in the log. Therefore, it's ok to put the new intents at the start of the dfops chain.
For the first example, the chains look like this:
t2: d4(t1), d5(t1), D1(t0), D2(t0), D3(t0)
t3: d5(t1), D1(t0), D2(t0), D3(t0)
...
t9: d9(t7), D3(t0)
t10: D3(t0)
t11: d10(t10), d11(t10)
t12: d11(t10)
For the second example, the chains look like this:
t1: D0(t0)
t2: d1(t1), d2(t1), D0'(t1)
t3: d2(t1), D0'(t1)
t4: D0'(t1)
t5: d1(t4), d2(t4), D0''(t4)
...
t148: D0<50 primes>(t147)
t149: d101(t148), d102(t148)
t150: d102(t148)
<done>
This actually sucks more for pinning the log tail (we try to roll to t10 while holding an intent item that was logged in t1) but we've solved problem #1. We've also reduced the maximum chain length from:
sum(all the new items) + nr_original_items
to:
max(new items that each original item creates) + nr_original_items
This solves problem #3 by sharply reducing the number of defer ops that can be attached to a transaction at any given time. The change makes the problem of log tail pinning worse, but is improvement we need to solve problem #2. Actually solving #2, however, is left to the next patch.
Note that a subsequent analysis of some hard-to-trigger reflink and COW livelocks on extremely fragmented filesystems (or systems running a lot of IO threads) showed the same symptoms -- uncomfortably large numbers of incore deferred work items and occasional stalls in the transaction grant code while waiting for log reservations. I think this patch and the next one will also solve these problems.
As originally written, the code used list_splice_tail_init instead of list_splice_init, so change that, and leave a short comment explaining our actions.
--- fs/xfs/libxfs/xfs_defer.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)
fs/xfs/libxfs/xfs_defer.c +10 -1
@@ -430,8 +430,17 @@ xfs_defer_finish_noroll(/* Until we run out of pending work to finish... */ while (!list_empty (&dop_pending) || !list_empty (&(*tp)->t_dfops)) { +/* +* Deferred items that are created in the process of finishing +* other deferred work items should be queued at the head of +* the pending list, which puts them ahead of the deferred work +* that was created by the caller. This keeps the number of +* pending work items to a minimum, which decreases the amount +* of time that any one intent item can stick around in memory, +* pinning the log tail. +*/ xfs_defer_create_intents (*tp); -list_splice_tail_init (&(*tp)->t_dfops, &dop_pending); +list_splice_init (&(*tp)->t_dfops, &dop_pending); error =xfs_defer_trans_roll (tp);if (error)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 28/147 xfs: fix realtime bitmap/summary file truncation when growing rt volume
In reply to Sasha LevinFrom: "Darrick J. Wong" <[email protected]>
[ Upstream commit f4c32e87de7d66074d5612567c5eac7325024428 ]
The realtime bitmap and summary files are regular files that are hidden away from the directory tree. Since they're regular files, inode inactivation will try to purge what it thinks are speculative preallocations beyond the incore size of the file. Unfortunately, xfs_growfs_rt forgets to update the incore size when it resizes the inodes, with the result that inactivating the rt inodes at unmount time will cause their contents to be truncated.
Fix this by updating the incore size when we change the ondisk size as part of updating the superblock. Note that we don't do this when we're allocating blocks to the rt inodes because we actually want those blocks to get purged if the growfs fails.
This fixes corruption complaints from the online rtsummary checker when running xfs/233. Since that test requires rmap, one can also trigger this by growing an rt volume, cycling the mount, and creating rt files.
--- fs/xfs/xfs_rtalloc.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
fs/xfs/xfs_rtalloc.c +8 -2
@@ -1016,10 +1016,13 @@ xfs_growfs_rt(xfs_ilock (mp->m_rbmip,XFS_ILOCK_EXCL );xfs_trans_ijoin (tp, mp->m_rbmip,XFS_ILOCK_EXCL );/* -* Update the bitmap inode's size. +* Update the bitmap inode's size ondisk and incore. We need +* to update the incore size so that inode inactivation won't +* punch what it thinks are "posteof" blocks. */ mp->m_rbmip->i_d.di_size = nsbp->sb_rbmblocks * nsbp->sb_blocksize; +i_size_write (VFS_I (mp->m_rbmip), mp->m_rbmip->i_d.di_size);xfs_trans_log_inode (tp, mp->m_rbmip,XFS_ILOG_CORE );/* * Get the summary inode into the transaction.
@@ -1027,9 +1030,12 @@ xfs_growfs_rt(xfs_ilock (mp->m_rsumip,XFS_ILOCK_EXCL );xfs_trans_ijoin (tp, mp->m_rsumip,XFS_ILOCK_EXCL );/* -* Update the summary inode's size. +* Update the summary inode's size. We need to update the +* incore size so that inode inactivation won't punch what it +* thinks are "posteof" blocks. */ mp->m_rsumip->i_d.di_size = nmp->m_rsumsize; +i_size_write (VFS_I (mp->m_rsumip), mp->m_rsumip->i_d.di_size);xfs_trans_log_inode (tp, mp->m_rsumip,XFS_ILOG_CORE );/* * Copy summary data from old to new sizes.
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 29/147 io_uring: don't set COMP_LOCKED if won't put
In reply to Sasha LevinFrom: Pavel Begunkov <[email protected]>
[ Upstream commit 368c5481ae7c6a9719c40984faea35480d9f4872 ]
__io_kill_linked_timeout() sets REQ_F_COMP_LOCKED for a linked timeout even if it can't cancel it, e.g. it's already running. It not only races with io_link_timeout_fn() for ->flags field, but also leaves the flag set and so io_link_timeout_fn() may find it and decide that it holds the lock. Hopefully, the second problem is potential.
--- fs/io_uring.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/io_uring.c +1 -1
@@ -1614,6 +1614,7 @@ static bool io_link_cancel_timeout(struct io_kiocb *req) ret =hrtimer_try_to_cancel (&req->io->timeout.timer);if (ret !=-1 ) { + req->flags |=REQ_F_COMP_LOCKED ;io_cqring_fill_event (req, -ECANCELED );io_commit_cqring (ctx); req->flags &= ~REQ_F_LINK_HEAD ;
@@ -1636,7 +1637,6 @@ static bool __io_kill_linked_timeout(struct io_kiocb *req)return false;list_del_init (&link->link_list); - link->flags |=REQ_F_COMP_LOCKED ; wake_ev =io_link_cancel_timeout (link); req->flags &= ~REQ_F_LINK_TIMEOUT ;return wake_ev;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 30/147 ath10k: fix retry packets update in station dump
In reply to Sasha LevinFrom: Venkateswara Naralasetty <[email protected]>
[ Upstream commit 67b927f9820847d30e97510b2f00cd142b9559b6 ]
When tx status enabled, retry count is updated from tx completion status. which is not working as expected due to firmware limitation where firmware can not provide per MSDU rate statistics from tx completion status. Due to this tx retry count is always 0 in station dump.
Fix this issue by updating the retry packet count from per peer statistics. This patch will not break on SDIO devices since, this retry count is already updating from peer statistics for SDIO devices.
Tested-on: QCA9984 PCI 10.4-3.6-00104
Tested-on: QCA9882 PCI 10.2.4-1.0-00047
--- drivers/net/wireless/ath/ath10k/htt_rx.c | 8 +++++--- drivers/net/wireless/ath/ath10k/mac.c | 5 +++-- 2 files changed, 8 insertions(+), 5 deletions(-)
drivers/net/wireless/ath/ath10k/htt_rx.c +5 -3
@@ -3575,12 +3575,14 @@ ath10k_update_per_peer_tx_stats(struct ath10k *ar, }if (ar->htt.disable_tx_comp) { - arsta->tx_retries += peer_stats->retry_pkts; arsta->tx_failed += peer_stats->failed_pkts; -ath10k_dbg (ar,ATH10K_DBG_HTT ,"htt tx retries %d tx failed %d\n" , - arsta->tx_retries, arsta->tx_failed); +ath10k_dbg (ar,ATH10K_DBG_HTT ,"tx failed %d\n" , + arsta->tx_failed); } + arsta->tx_retries += peer_stats->retry_pkts; +ath10k_dbg (ar,ATH10K_DBG_HTT ,"htt tx retries %d" , arsta->tx_retries); +if (ath10k_debug_is_extd_tx_stats_enabled (ar))ath10k_accumulate_per_peer_tx_stats (ar, arsta, peer_stats, rate_idx);
drivers/net/wireless/ath/ath10k/mac.c +3 -2
@@ -8542,12 +8542,13 @@ static void ath10k_sta_statistics(struct ieee80211_hw *hw, sinfo->filled |=BIT_ULL (NL80211_STA_INFO_TX_BITRATE );if (ar->htt.disable_tx_comp) { - sinfo->tx_retries = arsta->tx_retries; - sinfo->filled |=BIT_ULL (NL80211_STA_INFO_TX_RETRIES ); sinfo->tx_failed = arsta->tx_failed; sinfo->filled |=BIT_ULL (NL80211_STA_INFO_TX_FAILED ); } + sinfo->tx_retries = arsta->tx_retries; + sinfo->filled |=BIT_ULL (NL80211_STA_INFO_TX_RETRIES ); +ath10k_mac_sta_get_peer_stats_info (ar, sta, sinfo); }
Signature · 7 lines
2.25.1 _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Sasha Levin
maintainerPATCH 31/147 x86/kaslr: Initialize mem_limit to the real maximum address
In reply to Sasha LevinFrom: Arvind Sankar <[email protected]>
[ Upstream commit 451286940d95778e83fa7f97006316d995b4c4a8 ]
On 64-bit, the kernel must be placed below MAXMEM (64TiB with 4-level paging or 4PiB with 5-level paging). This is currently not enforced by KASLR, which thus implicitly relies on physical memory being limited to less than 64TiB.
On 32-bit, the limit is KERNEL_IMAGE_SIZE (512MiB). This is enforced by special checks in __process_mem_region().
Initialize mem_limit to the maximum (depending on architecture), instead of ULLONG_MAX, and make sure the command-line arguments can only decrease it. This makes the enforcement explicit on 64-bit, and eliminates the 32-bit specific checks to keep the kernel below 512M.
Check upfront to make sure the minimum address is below the limit before doing any work.
--- arch/x86/boot/compressed/kaslr.c | 41 +++++++++++++++++--------------- 1 file changed, 22 insertions(+), 19 deletions(-)
arch/x86/boot/compressed/kaslr.c +22 -19
@@ -87,8 +87,11 @@ static unsigned long get_boot_seed(void)static bool memmap_too_large; -/* Store memory limit specified by "mem=nn[KMG]" or "memmap=nn[KMG]" */ -static unsigned long long mem_limit =ULLONG_MAX ; +/* +* Store memory limit: MAXMEM on 64-bit and KERNEL_IMAGE_SIZE on 32-bit. +* It may be reduced by "mem=nn[KMG]" or "memmap=nn[KMG]" command line options. +*/ +static unsigned long long mem_limit;/* Number of immovable memory regions */ static int num_immovable_mem;
@@ -214,7 +217,7 @@ static void mem_avoid_memmap(enum parse_mode mode, char *str)if (start ==0 ) {/* Store the specified memory limit if size > 0 */ -if (size >0 ) +if (size >0 && size < mem_limit) mem_limit = size;continue ;
@@ -302,7 +305,8 @@ static void handle_mem_options(void)if (mem_size ==0 ) goto out; - mem_limit = mem_size; +if (mem_size < mem_limit) + mem_limit = mem_size; }else if (!strcmp (param,"efi_fake_mem" )) {mem_avoid_memmap (PARSE_EFI , val); }
@@ -314,7 +318,9 @@ static void handle_mem_options(void) }/* -* In theory, KASLR can put the kernel anywhere in the range of [16M, 64T). +* In theory, KASLR can put the kernel anywhere in the range of [16M, MAXMEM) +* on 64-bit, and [16M, KERNEL_IMAGE_SIZE) on 32-bit. +* * The mem_avoid array is used to store the ranges that need to be avoided * when KASLR searches for an appropriate random address. We must avoid any * regions that are unsafe to overlap with during decompression, and other
@@ -614,10 +620,6 @@ static void __process_mem_region(struct mem_vector *entry,unsigned long start_orig, end;struct mem_vector cur_entry; -/* On 32-bit, ignore entries entirely above our maximum. */ -if (IS_ENABLED (CONFIG_X86_32 ) && entry->start >=KERNEL_IMAGE_SIZE ) -return ; -/* Ignore entries entirely below our minimum. */ if (entry->start + entry->size < minimum)return ;
@@ -650,11 +652,6 @@ static void __process_mem_region(struct mem_vector *entry,/* Reduce size by any delta from the original address. */ region.size -= region.start - start_orig; -/* On 32-bit, reduce region size to fit within max size. */ -if (IS_ENABLED (CONFIG_X86_32 ) && - region.start + region.size >KERNEL_IMAGE_SIZE ) - region.size =KERNEL_IMAGE_SIZE - region.start; -/* Return if region can't contain decompressed kernel */ if (region.size < image_size)return ;
@@ -839,15 +836,16 @@ static void process_e820_entries(unsigned long minimum,static unsigned long find_random_phys_addr (unsigned long minimum,unsigned long image_size) { +/* Bail out early if it's impossible to succeed. */ +if (minimum + image_size > mem_limit) +return 0 ; +/* Check if we had too many memmaps. */ if (memmap_too_large) {debug_putstr ("Aborted memory entries scan (more than 4 memmap= args)!\n" );return 0 ; } -/* Make sure minimum is aligned. */ - minimum =ALIGN (minimum,CONFIG_PHYSICAL_ALIGN ); -if (process_efi_entries (minimum, image_size))return slots_fetch_random ();
@@ -860,8 +858,6 @@ static unsigned long find_random_virt_addr(unsigned long minimum, {unsigned long slots, random_addr; -/* Make sure minimum is aligned. */ - minimum =ALIGN (minimum,CONFIG_PHYSICAL_ALIGN );/* Align image_size for easy slot calculations. */ image_size =ALIGN (image_size,CONFIG_PHYSICAL_ALIGN );
@@ -908,6 +904,11 @@ void choose_random_location(unsigned long input,/* Prepare to add new identity pagetables on demand. */ initialize_identity_maps (); +if (IS_ENABLED (CONFIG_X86_32 )) + mem_limit =KERNEL_IMAGE_SIZE ; +else + mem_limit =MAXMEM ; +/* Record the various known unsafe memory ranges. */ mem_avoid_init (input, input_size, *output);
@@ -917,6 +918,8 @@ void choose_random_location(unsigned long input,* location: */ min_addr =min (*output,512UL <<20 ); +/* Make sure minimum is aligned. */ + min_addr =ALIGN (min_addr,CONFIG_PHYSICAL_ALIGN );/* Walk available memory entries to find a random address. */ random_addr =find_random_phys_addr (min_addr, output_size);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 32/147 drm/ast: Separate DRM driver from PCI code
In reply to Sasha LevinFrom: Thomas Zimmermann <[email protected]>
[ Upstream commit d50ace1e72f05708cc5dbc89b9bbb9873f150092 ]
Putting the DRM driver to the top of the file and the PCI code to the bottom makes ast_drv.c more readable. While at it, the patch prefixes file-scope variables with ast_.
--- drivers/gpu/drm/ast/ast_drv.c | 59 ++++++++++++++++++----------------- 1 file changed, 31 insertions(+), 28 deletions(-)
drivers/gpu/drm/ast/ast_drv.c +31 -28
@@ -43,9 +43,33 @@ int ast_modeset = -1;MODULE_PARM_DESC (modeset,"Disable/Enable modesetting" );module_param_named (modeset, ast_modeset, int,0400 ); -#define PCI_VENDOR_ASPEED 0x1a03 +/* +* DRM driver +*/ + +DEFINE_DRM_GEM_FOPS (ast_fops); + +static struct drm_driver ast_driver = { + .driver_features =DRIVER_ATOMIC | +DRIVER_GEM | +DRIVER_MODESET , + + .fops = &ast_fops, + .name =DRIVER_NAME , + .desc =DRIVER_DESC , + .date =DRIVER_DATE , + .major =DRIVER_MAJOR , + .minor =DRIVER_MINOR , + .patchlevel =DRIVER_PATCHLEVEL , -static struct drm_driver driver; +DRM_GEM_VRAM_DRIVER +}; + +/* +* PCI driver +*/ + +#define PCI_VENDOR_ASPEED 0x1a03#define AST_VGA_DEVICE (id, info) { \ .class = PCI_BASE_CLASS_DISPLAY << 16, \
@@ -56,13 +80,13 @@ static struct drm_driver driver; .subdevice = PCI_ANY_ID, \ .driver_data = (unsigned long) info } -static const struct pci_device_id pciidlist[] = { +static const struct pci_device_id ast_pciidlist[] = {AST_VGA_DEVICE (PCI_CHIP_AST2000 ,NULL ),AST_VGA_DEVICE (PCI_CHIP_AST2100 ,NULL ), {0 ,0 ,0 }, }; -MODULE_DEVICE_TABLE (pci, pciidlist); +MODULE_DEVICE_TABLE (pci, ast_pciidlist);static void ast_kick_out_firmware_fb (struct pci_dev *pdev) {
@@ -94,7 +118,7 @@ static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)if (ret)return ret; - dev =drm_dev_alloc (&driver, &pdev->dev); + dev =drm_dev_alloc (&ast_driver, &pdev->dev);if (IS_ERR (dev))return PTR_ERR (dev);
@@ -118,11 +142,9 @@ static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) err_drm_dev_put:drm_dev_put (dev);return ret; - } -static void -ast_pci_remove (struct pci_dev *pdev) +static void ast_pci_remove (struct pci_dev *pdev) {struct drm_device *dev =pci_get_drvdata (pdev);
@@ -217,30 +239,12 @@ static const struct dev_pm_ops ast_pm_ops = {static struct pci_driver ast_pci_driver = { .name =DRIVER_NAME , - .id_table = pciidlist, + .id_table = ast_pciidlist, .probe = ast_pci_probe, .remove = ast_pci_remove, .driver.pm = &ast_pm_ops, }; -DEFINE_DRM_GEM_FOPS (ast_fops); - -static struct drm_driver driver = { - .driver_features =DRIVER_ATOMIC | -DRIVER_GEM | -DRIVER_MODESET , - - .fops = &ast_fops, - .name =DRIVER_NAME , - .desc =DRIVER_DESC , - .date =DRIVER_DATE , - .major =DRIVER_MAJOR , - .minor =DRIVER_MINOR , - .patchlevel =DRIVER_PATCHLEVEL , - -DRM_GEM_VRAM_DRIVER -}; -static int __init ast_init (void ) {if (vgacon_text_force () && ast_modeset ==-1 )
@@ -261,4 +265,3 @@ module_exit(ast_exit);MODULE_AUTHOR (DRIVER_AUTHOR );MODULE_DESCRIPTION (DRIVER_DESC );MODULE_LICENSE ("GPL and additional rights" ); -
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 33/147 drm/amdgpu: restore ras flags when user resets eeprom(v2)
In reply to Sasha LevinFrom: Guchun Chen <[email protected]>
[ Upstream commit bf0b91b78f002faa1be1902a75eeb0797f9fbcf3 ]
RAS flags needs to be cleaned as well when user requires one clean eeprom.
v2: RAS flags shall be restored after eeprom reset succeeds.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +10 -3
@@ -367,12 +367,19 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *static ssize_t amdgpu_ras_debugfs_eeprom_write (struct file *f,const char __user *buf,size_t size,loff_t *pos) { -struct amdgpu_device *adev = (struct amdgpu_device *)file_inode (f)->i_private; +struct amdgpu_device *adev = + (struct amdgpu_device *)file_inode (f)->i_private;int ret; - ret =amdgpu_ras_eeprom_reset_table (&adev->psp.ras.ras->eeprom_control); + ret =amdgpu_ras_eeprom_reset_table ( + &(amdgpu_ras_get_context (adev)->eeprom_control)); -return ret ==1 ? size : -EIO ; +if (ret ==1 ) { +amdgpu_ras_get_context (adev)->flags =RAS_DEFAULT_FLAGS ; +return size; + }else { +return -EIO ; + } }static const struct file_operations amdgpu_ras_debugfs_ctrl_ops = {
Signature · 6 lines
2.25.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 34/147 video: fbdev: pvr2fb: initialize variables
In reply to Sasha LevinFrom: Tom Rix <[email protected]>
[ Upstream commit 8e1ba47c60bcd325fdd097cd76054639155e5d2e ]
clang static analysis reports this repesentative error
pvr2fb.c:1049:2: warning: 1st function call argument
is an uninitialized value [core.CallAndMessage]
if (*cable_arg)
^~~~~~~~~~~~~~~Problem is that cable_arg depends on the input loop to set the cable_arg[0]. If it does not, then some random value from the stack is used.
A similar problem exists for output_arg.
So initialize cable_arg and output_arg.
--- drivers/video/fbdev/pvr2fb.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/video/fbdev/pvr2fb.c +2 -0
@@ -1028,6 +1028,8 @@ static int __init pvr2fb_setup(char *options)if (!options || !*options)return 0 ; + cable_arg[0 ] = output_arg[0 ] =0 ; +while ((this_opt =strsep (&options,"," ))) {if (!*this_opt)continue ;
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 35/147 ath10k: start recovery process when payload length exceeds max htc length for sdio
In reply to Sasha LevinFrom: Wen Gong <[email protected]>
[ Upstream commit 2fd3c8f34d08af0a6236085f9961866ad92ef9ec ]
When simulate random transfer fail for sdio write and read, it happened "payload length exceeds max htc length" and recovery later sometimes.
Test steps:
1. Add config and update kernel:CONFIG_FAIL_MMC_REQUEST=yCONFIG_FAULT_INJECTION=yCONFIG_FAULT_INJECTION_DEBUG_FS=y
2. Run simulate fail:
cd /sys/kernel/debug/mmc1/fail_mmc_request
echo 10 > probability
echo 10 > times # repeat until hitting issues
3. It happened payload length exceeds max htc length.
[ 199.935506] ath10k_sdio mmc1:0001:1: payload length 57005 exceeds max htc length: 4088 .... [ 264.990191] ath10k_sdio mmc1:0001:1: payload length 57005 exceeds max htc length: 4088
4. after some time, such as 60 seconds, it start recovery which triggered by wmi command timeout for periodic scan.
[ 269.229232] ieee80211 phy0: Hardware restart was requested [ 269.734693] ath10k_sdio mmc1:0001:1: device successfully recovered
The simulate fail of sdio is not a real sdio transter fail, it only set an error status in mmc_should_fail_request after the transfer end, actually the transfer is success, then sdio_io_rw_ext_helper will return error status and stop transfer the left data. For example, the really RX len is 286 bytes, then it will split to 2 blocks in sdio_io_rw_ext_helper, one is 256 bytes, left is 30 bytes, if the first 256 bytes get an error status by mmc_should_fail_request,then the left 30 bytes will not read in this RX operation. Then when the next RX arrive, the left 30 bytes will be considered as the header of the read, the top 4 bytes of the 30 bytes will be considered as lookaheads, but actually the 4 bytes is not the lookaheads, so the len from this lookaheads is not correct, it exceeds max htc length 4088 sometimes. When happened exceeds, the buffer chain is not matched between firmware and ath10k, then it need to start recovery ASAP. Recently then recovery will be started by wmi command timeout, but it will be long time later, for example, it is 60+ seconds later from the periodic scan, if it does not have periodic scan, it will be longer.
Start recovery when it happened "payload length exceeds max htc length" will be reasonable.
This patch only effect sdio chips.
Tested with QCA6174 SDIO with firmware WLAN.RMH.4.4.1-00029.
--- drivers/net/wireless/ath/ath10k/sdio.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/net/wireless/ath/ath10k/sdio.c +4 -0
@@ -557,6 +557,10 @@ static int ath10k_sdio_mbox_rx_alloc(struct ath10k *ar,le16_to_cpu (htc_hdr->len), ATH10K_HTC_MBOX_MAX_PAYLOAD_LENGTH); ret = -ENOMEM ; + +queue_work (ar->workqueue, &ar->restart_work); +ath10k_warn (ar,"exceeds length, start recovery\n" ); + goto err; }
Signature · 7 lines
2.25.1 _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Sasha Levin
maintainerPATCH 36/147 ath10k: fix VHT NSS calculation when STBC is enabled
In reply to Sasha LevinFrom: Sathishkumar Muruganandam <[email protected]>
[ Upstream commit 99f41b8e43b8b4b31262adb8ac3e69088fff1289 ]
When STBC is enabled, NSTS_SU value need to be accounted for VHT NSS calculation for SU case.
Without this fix, 1SS + STBC enabled case was reported wrongly as 2SS in radiotap header on monitor mode capture.
Tested-on: QCA9984 10.4-3.10-00047
--- drivers/net/wireless/ath/ath10k/htt_rx.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/net/wireless/ath/ath10k/htt_rx.c +7 -1
@@ -941,6 +941,7 @@ static void ath10k_htt_rx_h_rates(struct ath10k *ar,u8 preamble =0 ;u8 group_id;u32 info1, info2, info3; +u32 stbc, nsts_su; info1 =__le32_to_cpu (rxd->ppdu_start.info1); info2 =__le32_to_cpu (rxd->ppdu_start.info2);
@@ -985,11 +986,16 @@ static void ath10k_htt_rx_h_rates(struct ath10k *ar,*/ bw = info2 &3 ; sgi = info3 &1 ; + stbc = (info2 >>3 ) &1 ; group_id = (info2 >>4 ) &0x3F ;if (GROUP_ID_IS_SU_MIMO (group_id)) { mcs = (info3 >>4 ) &0x0F ; - nss = ((info2 >>10 ) &0x07 ) +1 ; + nsts_su = ((info2 >>10 ) &0x07 ); +if (stbc) + nss = (nsts_su >>2 ) +1 ; +else + nss = (nsts_su +1 ); }else {/* Hardware doesn't decode VHT-SIG-B into Rx descriptor * so it's impossible to decode MCS. Also since
Signature · 7 lines
2.25.1 _______________________________________________ ath10k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath10k
Sasha Levin
maintainerPATCH 37/147 drm/scheduler: Scheduler priority fixes (v2)
In reply to Sasha LevinFrom: Luben Tuikov <[email protected]>
[ Upstream commit e2d732fdb7a9e421720a644580cd6a9400f97f60 ]
Remove DRM_SCHED_PRIORITY_LOW, as it was used in only one place.
Rename and separate by a line DRM_SCHED_PRIORITY_MAX to DRM_SCHED_PRIORITY_COUNT as it represents a (total) count of said priorities and it is used as such in loops throughout the code. (0-based indexing is the the count number.)
Remove redundant word HIGH in priority names, and rename *KERNEL* to *HIGH*, as it really means that, high.
v2: Add back KERNEL and remove SW and HW,
in lieu of a single HIGH between NORMAL and KERNEL.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 4 ++-- drivers/gpu/drm/amd/amdgpu/amdgpu_job.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c | 6 +++--- drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 2 +- drivers/gpu/drm/scheduler/sched_main.c | 4 ++-- include/drm/gpu_scheduler.h | 12 +++++++----- 8 files changed, 18 insertions(+), 16 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +2 -2
@@ -46,7 +46,7 @@ const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM] = {static int amdgpu_ctx_priority_permit (struct drm_file *filp,enum drm_sched_priority priority) { -if (priority <0 || priority >=DRM_SCHED_PRIORITY_MAX ) +if (priority <0 || priority >=DRM_SCHED_PRIORITY_COUNT )return -EINVAL ;/* NORMAL and below are accessible by everyone */
@@ -65,7 +65,7 @@ static int amdgpu_ctx_priority_permit(struct drm_file *filp,static enum gfx_pipe_priority amdgpu_ctx_sched_prio_to_compute_prio (enum drm_sched_priority prio) {switch (prio) { -case DRM_SCHED_PRIORITY_HIGH_HW : +case DRM_SCHED_PRIORITY_HIGH :case DRM_SCHED_PRIORITY_KERNEL :return AMDGPU_GFX_PIPE_PRIO_HIGH ;default :
drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +1 -1
@@ -251,7 +251,7 @@ void amdgpu_job_stop_all_jobs_on_sched(struct drm_gpu_scheduler *sched)int i;/* Signal all jobs not yet scheduled */ -for (i =DRM_SCHED_PRIORITY_MAX -1 ; i >=DRM_SCHED_PRIORITY_MIN ; i--) { +for (i =DRM_SCHED_PRIORITY_COUNT -1 ; i >=DRM_SCHED_PRIORITY_MIN ; i--) {struct drm_sched_rq *rq = &sched->sched_rq[i];if (!rq)
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +1 -1
@@ -267,7 +267,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring, &ring->sched; } -for (i =0 ; i <DRM_SCHED_PRIORITY_MAX ; ++i) +for (i =DRM_SCHED_PRIORITY_MIN ; i <DRM_SCHED_PRIORITY_COUNT ; ++i)atomic_set (&ring->num_jobs[i],0 );return 0 ;
drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +1 -1
@@ -243,7 +243,7 @@ struct amdgpu_ring {bool has_compute_vm_bug;bool no_scheduler; -atomic_t num_jobs[DRM_SCHED_PRIORITY_MAX ]; +atomic_t num_jobs[DRM_SCHED_PRIORITY_COUNT ];struct mutex priority_mutex;/* protected by priority_mutex */ int priority;
drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c +3 -3
@@ -36,14 +36,14 @@ enum drm_sched_priority amdgpu_to_sched_priority(int amdgpu_priority) {switch (amdgpu_priority) {case AMDGPU_CTX_PRIORITY_VERY_HIGH : -return DRM_SCHED_PRIORITY_HIGH_HW ; +return DRM_SCHED_PRIORITY_HIGH ;case AMDGPU_CTX_PRIORITY_HIGH : -return DRM_SCHED_PRIORITY_HIGH_SW ; +return DRM_SCHED_PRIORITY_HIGH ;case AMDGPU_CTX_PRIORITY_NORMAL :return DRM_SCHED_PRIORITY_NORMAL ;case AMDGPU_CTX_PRIORITY_LOW :case AMDGPU_CTX_PRIORITY_VERY_LOW : -return DRM_SCHED_PRIORITY_LOW ; +return DRM_SCHED_PRIORITY_MIN ;case AMDGPU_CTX_PRIORITY_UNSET :return DRM_SCHED_PRIORITY_UNSET ;default :
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +1 -1
@@ -2101,7 +2101,7 @@ void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev, bool enable) ring = adev->mman.buffer_funcs_ring; sched = &ring->sched; r =drm_sched_entity_init (&adev->mman.entity, -DRM_SCHED_PRIORITY_KERNEL , &sched, +DRM_SCHED_PRIORITY_KERNEL , &sched,1 ,NULL );if (r) { DRM_ERROR("Failed setting up TTM BO move entity (%d)\n",
drivers/gpu/drm/scheduler/sched_main.c +2 -2
@@ -625,7 +625,7 @@ drm_sched_select_entity(struct drm_gpu_scheduler *sched)return NULL ;/* Kernel run queue has higher priority than normal run queue*/ -for (i =DRM_SCHED_PRIORITY_MAX -1 ; i >=DRM_SCHED_PRIORITY_MIN ; i--) { +for (i =DRM_SCHED_PRIORITY_COUNT -1 ; i >=DRM_SCHED_PRIORITY_MIN ; i--) { entity =drm_sched_rq_select_entity (&sched->sched_rq[i]);if (entity)break ;
@@ -852,7 +852,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched, sched->name = name; sched->timeout = timeout; sched->hang_limit = hang_limit; -for (i =DRM_SCHED_PRIORITY_MIN ; i <DRM_SCHED_PRIORITY_MAX ; i++) +for (i =DRM_SCHED_PRIORITY_MIN ; i <DRM_SCHED_PRIORITY_COUNT ; i++)drm_sched_rq_init (sched, &sched->sched_rq[i]);init_waitqueue_head (&sched->wake_up_worker);
include/drm/gpu_scheduler.h +7 -5
@@ -33,14 +33,16 @@struct drm_gpu_scheduler ;struct drm_sched_rq ; +/* These are often used as an (initial) index +* to an array, and as such should start at 0. +*/ enum drm_sched_priority {DRM_SCHED_PRIORITY_MIN , -DRM_SCHED_PRIORITY_LOW =DRM_SCHED_PRIORITY_MIN ,DRM_SCHED_PRIORITY_NORMAL , -DRM_SCHED_PRIORITY_HIGH_SW , -DRM_SCHED_PRIORITY_HIGH_HW , +DRM_SCHED_PRIORITY_HIGH ,DRM_SCHED_PRIORITY_KERNEL , -DRM_SCHED_PRIORITY_MAX , + +DRM_SCHED_PRIORITY_COUNT ,DRM_SCHED_PRIORITY_INVALID =-1 ,DRM_SCHED_PRIORITY_UNSET =-2 };
@@ -274,7 +276,7 @@ struct drm_gpu_scheduler {uint32_t hw_submission_limit;long timeout;const char *name; -struct drm_sched_rq sched_rq[DRM_SCHED_PRIORITY_MAX ]; +struct drm_sched_rq sched_rq[DRM_SCHED_PRIORITY_COUNT ];wait_queue_head_t wake_up_worker;wait_queue_head_t job_scheduled;atomic_t hw_rq_count;
Signature · 6 lines
2.25.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 38/147 drm/brige/megachips: Add checking if ge_b850v3_lvds_init() is working correctly
In reply to Sasha LevinFrom: Nadezda Lutovinova <[email protected]>
[ Upstream commit f688a345f0d7a6df4dd2aeca8e4f3c05e123a0ee ]
If ge_b850v3_lvds_init() does not allocate memory for ge_b850v3_lvds_ptr, then a null pointer dereference is accessed.
The patch adds checking of the return value of ge_b850v3_lvds_init().
Found by Linux Driver Verification project (linuxtesting.org).
--- .../gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)
drivers/gpu/drm/bridge/megachips-stdpxxxx-ge-b850v3-fw.c +10 -2
@@ -302,8 +302,12 @@ static int stdp4028_ge_b850v3_fw_probe(struct i2c_client *stdp4028_i2c, const struct i2c_device_id *id) {struct device *dev = &stdp4028_i2c->dev; +int ret; + + ret =ge_b850v3_lvds_init (dev); -ge_b850v3_lvds_init (dev); +if (ret) +return ret; ge_b850v3_lvds_ptr->stdp4028_i2c = stdp4028_i2c;i2c_set_clientdata (stdp4028_i2c, ge_b850v3_lvds_ptr);
@@ -361,8 +365,12 @@ static int stdp2690_ge_b850v3_fw_probe(struct i2c_client *stdp2690_i2c, const struct i2c_device_id *id) {struct device *dev = &stdp2690_i2c->dev; +int ret; + + ret =ge_b850v3_lvds_init (dev); -ge_b850v3_lvds_init (dev); +if (ret) +return ret; ge_b850v3_lvds_ptr->stdp2690_i2c = stdp2690_i2c;i2c_set_clientdata (stdp2690_i2c, ge_b850v3_lvds_ptr);
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 39/147 ASoC: SOF: fix a runtime pm issue in SOF when HDMI codec doesn't work
In reply to Sasha LevinFrom: Rander Wang <[email protected]>
[ Upstream commit 6c63c954e1c52f1262f986f36d95f557c6f8fa94 ]
When hda_codec_probe() doesn't initialize audio component, we disable the codec and keep going. However,the resources are not released. The child_count of SOF device is increased in snd_hdac_ext_bus_device_init but is not decrease in error case, so SOF can't get suspended.
snd_hdac_ext_bus_device_exit will be invoked in HDA framework if it gets a error. Now copy this behavior to release resources and decrease SOF device child_count to release SOF device.
--- sound/soc/sof/intel/hda-codec.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
sound/soc/sof/intel/hda-codec.c +2 -2
@@ -151,7 +151,7 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,if (!hdev->bus->audio_component) {dev_dbg (sdev->dev,"iDisp hw present but no driver\n" ); -return -ENOENT ; + goto error; } hda_priv->need_display_power = true; }
@@ -174,7 +174,7 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,* other return codes without modification */ if (ret ==0 ) - ret = -ENOENT ; + goto error; }return ret;
Signature · 1 line
2.25.1
Build error with 5.9.5 in sound/soc/sof/intel/hda-codec.c (was: [PATCH AUTOSEL 5.9 039/147] ASoC: SOF: fix a runtime pm issue in SOF when HDMI codec doesn't work)
Lo! I just tried to compile 5.9.5 and ran into a build error with below patch. I only did a quick look (I have to leave the keyboard soon), but seems the patch quoted below that was added to 5.9.5 depends on 11ec0edc6408 (git.kernel.org/linus/11ec0edc6408) which wasn't backported.
The build error can be found here:
https://kojipkgs.fedoraproject.org//work/tasks/8246/54978246/build.log
Relevant part:
+ make -s 'HOSTCFLAGS=-O2 -flto=auto -ffat-lto-objects -fexceptions -g -grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -fcommon -m64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection' 'HOSTLDFLAGS=-Wl,-z,relro -Wl,--as-needed -Wl,-z,now -specs=/usr/lib/rpm/redhat/redhat-hardened-ld ' ARCH=x86_64 'KCFLAGS= ' WITH_GCOV=0 -j48 modules sound/soc/sof/intel/hda-codec.c: In function 'hda_codec_probe':sound/soc/sof/intel/hda-codec.c:177:4: error: label 'error' used but not defined
177 | goto error;
| ^~~~make[4]: *** [scripts/Makefile.build:283: sound/soc/sof/intel/hda-codec.o] Error 1 make[3]: *** [scripts/Makefile.build:500: sound/soc/sof/intel] Error 2 make[3]: *** Waiting for unfinished jobs....
make[2]: *** [scripts/Makefile.build:500: sound/soc/sof] Error 2 make[2]: *** Waiting for unfinished jobs....
make[1]: *** [scripts/Makefile.build:500: sound/soc] Error 2
make: *** [Makefile:1784: sound] Error 2
make: *** Waiting for unfinished jobs....
+ exit 1
Looks like the compiler is right from a quick look at
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/sound/soc/sof/intel/hda-codec.c?h=linux-5.9.y&id=43836fdc9e318a11233cf19c5ee7ffb04e8e5d8f
But as I said, I lack the time for a closer look.
Ciao, Thorsten
Am 27.10.20 um 00:47 schrieb Sasha Levin:
Quoted 48 lines
From: Rander Wang <[email protected]>
[ Upstream commit
6c63c954e1c52f1262f986f36d95f557c6f8fa94]When
hda_codec_probe()doesn't initialize audio component, we disable the codec and keep going. However,the resources are not released. Thechild_countof SOF device is increased insnd_hdac_ext_bus_device_initbut is not decrease in error case, so SOF can't get suspended.
snd_hdac_ext_bus_device_exitwill be invoked in HDA framework if it gets a error. Now copy this behavior to release resources and decrease SOF devicechild_countto release SOF device.Signed-off-by Rander Wang <[email protected]>Reviewed-by Pierre-Louis Bossart <[email protected]>Reviewed-by Bard Liao <[email protected]>Reviewed-by Guennadi Liakhovetski <[email protected]>Signed-off-by Ranjani Sridharan <[email protected]>Signed-off-by Mark Brown <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- sound/soc/sof/intel/hda-codec.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)sound/soc/sof/intel/hda-codec.c +2
-2@@ -151,7 +151,7 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,if (!hdev->bus->audio_component) {dev_dbg (sdev->dev,"iDisp hw present but no driver\n" ); -return -ENOENT ; + goto error; } hda_priv->need_display_power = true; }@@ -174,7 +174,7 @@ static int hda_codec_probe(struct snd_sof_dev *sdev, int address,* other return codes without modification */ if (ret ==0 ) - ret = -ENOENT ; + goto error; }return ret;
Greg KH
maintainerRe: Build error with 5.9.5 in sound/soc/sof/intel/hda-codec.c (was: [PATCH AUTOSEL 5.9 039/147] ASoC: SOF: fix a runtime pm issue in SOF when HDMI codec doesn't work)
Quoted 35 lines · Thorsten Leemhuis
Lo! I just tried to compile 5.9.5 and ran into a build error with below patch. I only did a quick look (I have to leave the keyboard soon), but seems the patch quoted below that was added to 5.9.5 depends on
11ec0edc6408(git.kernel.org/linus/11ec0edc6408) which wasn't backported.The build error can be found here:
https://kojipkgs.fedoraproject.org//work/tasks/8246/54978246/build.logRelevant part:
+ make -s 'HOSTCFLAGS=-O2 -flto=auto -ffat-lto-objects -fexceptions -g
(quoted diff) +0
-5-grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -fcommon -m64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection''HOSTLDFLAGS=-Wl,-z,relro -Wl,--as-needed -Wl,-z,now -specs=/usr/lib/rpm/redhat/redhat-hardened-ld ' ARCH=x86_64 'KCFLAGS= ' WITH_GCOV=0 -j48 modules
sound/soc/sof/intel/hda-codec.c: In function 'hda_codec_probe':sound/soc/sof/intel/hda-codec.c:177:4: error: label 'error' used but not defined177 | goto error; | ^~~~make[4]: *** [
scripts/Makefile.build:283:sound/soc/sof/intel/hda-codec.o] Error 1
make[3]: *** [scripts/Makefile.build:500:sound/soc/sof/intel] Error 2 make[3]: *** Waiting for unfinished jobs....
make[2]: *** [scripts/Makefile.build:500:sound/soc/sof] Error 2 make[2]: *** Waiting for unfinished jobs....
make[1]: *** [scripts/Makefile.build:500:sound/soc] Error 2
make: *** [Makefile:1784: sound] Error 2
make: *** Waiting for unfinished jobs....
+ exit 1
Quoted 4 lines
Looks like the compiler is right from a quick look at
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/sound/soc/sof/intel/hda-codec.c?h=linux-5.9.y&id=43836fdc9e318a11233cf19c5ee7ffb04e8e5d8fBut as I said, I lack the time for a closer look.
Thanks, people are looking at it already:
https://lore.kernel.org/r/[email protected]
Greg KH
maintainerRe: Build error with 5.9.5 in sound/soc/sof/intel/hda-codec.c (was: [PATCH AUTOSEL 5.9 039/147] ASoC: SOF: fix a runtime pm issue in SOF when HDMI codec doesn't work)
Quoted 36 lines · Greg KH
Quoted 35 lines · Thorsten Leemhuis
Lo! I just tried to compile 5.9.5 and ran into a build error with below patch. I only did a quick look (I have to leave the keyboard soon), but seems the patch quoted below that was added to 5.9.5 depends on
11ec0edc6408(git.kernel.org/linus/11ec0edc6408) which wasn't backported.The build error can be found here:
https://kojipkgs.fedoraproject.org//work/tasks/8246/54978246/build.logRelevant part:
+ make -s 'HOSTCFLAGS=-O2 -flto=auto -ffat-lto-objects -fexceptions -g
(quoted diff) +0
-5-grecord-gcc-switches -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -Wp,-D_GLIBCXX_ASSERTIONS -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 -fstack-protector-strong -specs=/usr/lib/rpm/redhat/redhat-annobin-cc1 -fcommon -m64 -mtune=generic -fasynchronous-unwind-tables -fstack-clash-protection''HOSTLDFLAGS=-Wl,-z,relro -Wl,--as-needed -Wl,-z,now -specs=/usr/lib/rpm/redhat/redhat-hardened-ld ' ARCH=x86_64 'KCFLAGS= ' WITH_GCOV=0 -j48 modules
sound/soc/sof/intel/hda-codec.c: In function 'hda_codec_probe':sound/soc/sof/intel/hda-codec.c:177:4: error: label 'error' used but not defined177 | goto error; | ^~~~make[4]: *** [
scripts/Makefile.build:283:sound/soc/sof/intel/hda-codec.o] Error 1
make[3]: *** [scripts/Makefile.build:500:sound/soc/sof/intel] Error 2 make[3]: *** Waiting for unfinished jobs....
make[2]: *** [scripts/Makefile.build:500:sound/soc/sof] Error 2 make[2]: *** Waiting for unfinished jobs....
make[1]: *** [scripts/Makefile.build:500:sound/soc] Error 2
make: *** [Makefile:1784: sound] Error 2
make: *** Waiting for unfinished jobs....
+ exit 1
Quoted 7 lines
Quoted 4 lines
Looks like the compiler is right from a quick look at
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/sound/soc/sof/intel/hda-codec.c?h=linux-5.9.y&id=43836fdc9e318a11233cf19c5ee7ffb04e8e5d8fBut as I said, I lack the time for a closer look.
Thanks, people are looking at it already:
https://lore.kernel.org/r/[email protected]
Should now be fixed in 5.9.6.
thanks,
greg k-h
Sasha Levin
maintainerPATCH 40/147 selftests/x86/fsgsbase: Reap a forgotten child
In reply to Sasha LevinFrom: Andy Lutomirski <[email protected]>
[ Upstream commit ab2dd173330a3f07142e68cd65682205036cd00f ]
The ptrace() test forgot to reap its child. Reap it.
--- tools/testing/selftests/x86/fsgsbase.c | 3 +++ 1 file changed, 3 insertions(+)
tools/testing/selftests/x86/fsgsbase.c +3 -0
@@ -517,6 +517,9 @@ static void test_ptrace_write_gsbase(void) END:ptrace (PTRACE_CONT , child,NULL ,NULL ); +wait (&status); +if (!WIFEXITED (status)) +printf ("[WARN]\tChild didn't exit cleanly.\n" ); }int main ()
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 41/147 drm/bridge_connector: Set default status connected for eDP connectors
In reply to Sasha LevinFrom: Enric Balletbo i Serra <[email protected]>
[ Upstream commit c5589b39549d1875bb506da473bf4580c959db8c ]
In an eDP application, HPD is not required and on most bridge chips useless. If HPD is not used, we need to set initial status as connected, otherwise the connector created by the drm_bridge_connector API remains in an unknown state.
--- drivers/gpu/drm/drm_bridge_connector.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/drm_bridge_connector.c +1 -0
@@ -187,6 +187,7 @@ drm_bridge_connector_detect(struct drm_connector *connector, bool force)case DRM_MODE_CONNECTOR_DPI :case DRM_MODE_CONNECTOR_LVDS :case DRM_MODE_CONNECTOR_DSI : +case DRM_MODE_CONNECTOR_eDP: status = connector_status_connected;break ;default :
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 42/147 media: videodev2.h: RGB BT2020 and HSV are always full range
In reply to Sasha LevinFrom: Hans Verkuil <[email protected]>
[ Upstream commit b305dfe2e93434b12d438434461b709641f62af4 ]
The default RGB quantization range for BT.2020 is full range (just as for all the other RGB pixel encodings), not limited range.
Update the V4L2_MAP_QUANTIZATION_DEFAULT macro and documentation accordingly.
Also mention that HSV is always full range and cannot be limited range.
When RGB BT2020 was introduced in V4L2 it was not clear whether it should be limited or full range, but full range is the right (and consistent) choice.
--- .../media/v4l/colorspaces-defs.rst | 9 ++++----- .../media/v4l/colorspaces-details.rst | 5 ++--- include/uapi/linux/videodev2.h | 17 ++++++++--------- 3 files changed, 14 insertions(+), 17 deletions(-)
Documentation/userspace-api/media/v4l/colorspaces-defs.rst +4 -5
@@ -36,8 +36,7 @@ whole range, 0-255, dividing the angular value by 1.41. The enum :c:type:`v4l2_hsv_encoding` specifies which encoding is used. .. note:: The default R'G'B' quantization is full range for all - colorspaces except for BT.2020 which uses limited range R'G'B' - quantization. + colorspaces. HSV formats are always full range. .. tabularcolumns:: |p{6.7cm}|p{10.8cm}|
@@ -169,8 +168,8 @@ whole range, 0-255, dividing the angular value by 1.41. The enum - Details * - ``V4L2_QUANTIZATION_DEFAULT`` - Use the default quantization encoding as defined by the - colorspace. This is always full range for R'G'B' (except for the - BT.2020 colorspace) and HSV. It is usually limited range for Y'CbCr. + colorspace. This is always full range for R'G'B' and HSV. + It is usually limited range for Y'CbCr. * - ``V4L2_QUANTIZATION_FULL_RANGE`` - Use the full range quantization encoding. I.e. the range [0…1] is mapped to [0…255] (with possible clipping to [1…254] to avoid the
@@ -180,4 +179,4 @@ whole range, 0-255, dividing the angular value by 1.41. The enum * - ``V4L2_QUANTIZATION_LIM_RANGE`` - Use the limited range quantization encoding. I.e. the range [0…1] is mapped to [16…235]. Cb and Cr are mapped from [-0.5…0.5] to - [16…240]. + [16…240]. Limited Range cannot be used with HSV.
Documentation/userspace-api/media/v4l/colorspaces-details.rst +2 -3
@@ -377,9 +377,8 @@ Colorspace BT.2020 (V4L2_COLORSPACE_BT2020) The :ref:`itu2020` standard defines the colorspace used by Ultra-high definition television (UHDTV). The default transfer function is ``V4L2_XFER_FUNC_709``. The default Y'CbCr encoding is -``V4L2_YCBCR_ENC_BT2020``. The default R'G'B' quantization is limited -range (!), and so is the default Y'CbCr quantization. The chromaticities -of the primary colors and the white reference are: +``V4L2_YCBCR_ENC_BT2020``. The default Y'CbCr quantization is limited range. +The chromaticities of the primary colors and the white reference are:
include/uapi/linux/videodev2.h +8 -9
@@ -373,9 +373,9 @@ enum v4l2_hsv_encoding {enum v4l2_quantization {/* -* The default for R'G'B' quantization is always full range, except -* for the BT2020 colorspace. For Y'CbCr the quantization is always -* limited range, except for COLORSPACE_JPEG: this is full range. +* The default for R'G'B' quantization is always full range. +* For Y'CbCr the quantization is always limited range, except +* for COLORSPACE_JPEG: this is full range. */ V4L2_QUANTIZATION_DEFAULT =0 ,V4L2_QUANTIZATION_FULL_RANGE =1 ,
@@ -384,14 +384,13 @@ enum v4l2_quantization {/* * Determine how QUANTIZATION_DEFAULT should map to a proper quantization. -* This depends on whether the image is RGB or not, the colorspace and the -* Y'CbCr encoding. +* This depends on whether the image is RGB or not, the colorspace. +* The Y'CbCr encoding is not used anymore, but is still there for backwards +* compatibility. */ #define V4L2_MAP_QUANTIZATION_DEFAULT (is_rgb_or_hsv, colsp, ycbcr_enc) \ - (((is_rgb_or_hsv) && (colsp) == V4L2_COLORSPACE_BT2020) ? \ - V4L2_QUANTIZATION_LIM_RANGE : \ - (((is_rgb_or_hsv) || (colsp) == V4L2_COLORSPACE_JPEG) ? \ - V4L2_QUANTIZATION_FULL_RANGE : V4L2_QUANTIZATION_LIM_RANGE)) + (((is_rgb_or_hsv) || (colsp) == V4L2_COLORSPACE_JPEG) ? \ + V4L2_QUANTIZATION_FULL_RANGE : V4L2_QUANTIZATION_LIM_RANGE)/* * Deprecated names for opRGB colorspace (IEC 61966-2-5)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 43/147 ASoC: AMD: Clean kernel log from deferred probe error messages
In reply to Sasha LevinFrom: Akshu Agrawal <[email protected]>
[ Upstream commit f7660445c8e7fda91e8b944128554249d886b1d4 ]
While the driver waits for DAIs to be probed and retries probing, have the error messages at debug level instead of error.
--- sound/soc/amd/acp3x-rt5682-max9836.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)
sound/soc/amd/acp3x-rt5682-max9836.c +8 -3
@@ -472,12 +472,17 @@ static int acp3x_probe(struct platform_device *pdev) ret =devm_snd_soc_register_card (&pdev->dev, card);if (ret) { -dev_err (&pdev->dev, +if (ret != -EPROBE_DEFER ) +dev_err (&pdev->dev,"devm_snd_soc_register_card(%s) failed: %d\n" , card->name, ret); -return ret; +else +dev_dbg (&pdev->dev, +"devm_snd_soc_register_card(%s) probe deferred: %d\n" , + card->name, ret); } -return 0 ; + +return ret; }static const struct acpi_device_id acp3x_audio_acpi_match[] = {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 44/147 misc: fastrpc: fix common struct sg_table related issues
In reply to Sasha LevinFrom: Marek Szyprowski <[email protected]>
[ Upstream commit 7cd7edb89437457ec36ffdbb970cc314d00c4aba ]
The Documentation/DMA-API-HOWTO.txt states that the dma_map_sg() function returns the number of the created entries in the DMA address space. However the subsequent calls to the dma_sync_sg_for_{device,cpu}() and dma_unmap_sg must be called with the original number of the entries passed to the dma_map_sg().
struct sg_table is a common structure used for describing a non-contiguous memory buffer, used commonly in the DRM and graphics subsystems. It consists of a scatterlist with memory pages and DMA addresses (sgl entry), as well as the number of scatterlist entries: CPU pages (orig_nents entry) and DMA mapped pages (nents entry).
It turned out that it was a common mistake to misuse nents and orig_nents entries, calling DMA-mapping functions with a wrong number of entries or ignoring the number of mapped entries returned by the dma_map_sg() function.
To avoid such issues, lets use a common dma-mapping wrappers operating directly on the struct sg_table objects and use scatterlist page iterators where possible. This, almost always, hides references to the nents and orig_nents entries, making the code robust, easier to follow and copy/paste safe.
--- drivers/misc/fastrpc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/misc/fastrpc.c +2 -2
@@ -518,7 +518,7 @@ fastrpc_map_dma_buf(struct dma_buf_attachment *attachment, table = &a->sgt; -if (!dma_map_sg (attachment->dev, table->sgl, table->nents, dir)) +if (!dma_map_sgtable (attachment->dev, table, dir,0 ))return ERR_PTR (-ENOMEM );return table;
@@ -528,7 +528,7 @@ static void fastrpc_unmap_dma_buf(struct dma_buf_attachment *attach,struct sg_table *table, enum dma_data_direction dir) { - dma_unmap_sg(attach->dev, table->sgl, table->nents, dir); + dma_unmap_sgtable(attach->dev, table, dir, 0); }static void fastrpc_release (struct dma_buf *dmabuf)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 45/147 staging: wfx: fix potential use before init
In reply to Sasha LevinFrom: Jérôme Pouiller <[email protected]>
[ Upstream commit ce3653a8d3db096aa163fc80239d8ec1305c81fa ]
The trace below can appear:
[83613.832200] INFO: trying to register non-static key.
[83613.837248] the code is fine but needs lockdep annotation.
[83613.842808] turning off the locking correctness validator.
[83613.848375] CPU: 3 PID: 141 Comm: kworker/3:2H Tainted: G O 5.6.13-silabs15 #2
[83613.857019] Hardware name: BCM2835
[83613.860605] Workqueue: events_highpri bh_work [wfx]
[83613.865552] Backtrace:
[83613.868041] [<c010f2cc>] (dump_backtrace) from [<c010f7b8>] (show_stack+0x20/0x24)
[83613.881463] [<c010f798>] (show_stack) from [<c0d82138>] (dump_stack+0xe8/0x114)
[83613.888882] [<c0d82050>] (dump_stack) from [<c01a02ec>] (register_lock_class+0x748/0x768)
[83613.905035] [<c019fba4>] (register_lock_class) from [<c019da04>] (__lock_acquire+0x88/0x13dc)
[83613.924192] [<c019d97c>] (__lock_acquire) from [<c019f6a4>] (lock_acquire+0xe8/0x274)
[83613.942644] [<c019f5bc>] (lock_acquire) from [<c0daa5dc>] (_raw_spin_lock_irqsave+0x58/0x6c)
[83613.961714] [<c0daa584>] (_raw_spin_lock_irqsave) from [<c0ab3248>] (skb_dequeue+0x24/0x78)
[83613.974967] [<c0ab3224>] (skb_dequeue) from [<bf330db0>] (wfx_tx_queues_get+0x96c/0x1294 [wfx])
[83613.989728] [<bf330444>] (wfx_tx_queues_get [wfx]) from [<bf320454>] (bh_work+0x454/0x26d8 [wfx])
[83614.009337] [<bf320000>] (bh_work [wfx]) from [<c014c920>] (process_one_work+0x23c/0x7ec)
[83614.028141] [<c014c6e4>] (process_one_work) from [<c014cf1c>] (worker_thread+0x4c/0x55c)
[83614.046861] [<c014ced0>] (worker_thread) from [<c0154c04>] (kthread+0x138/0x168)
[83614.064876] [<c0154acc>] (kthread) from [<c01010b4>] (ret_from_fork+0x14/0x20)
[83614.072200] Exception stack(0xecad3fb0 to 0xecad3ff8)
[83614.077323] 3fa0: 00000000 00000000 00000000 00000000
[83614.085620] 3fc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[83614.093914] 3fe0: 00000000 00000000 00000000 00000000 00000013 00000000Indeed, the code of wfx_add_interface() shows that the interface is enabled to early. So, the spinlock associated with some skb_queue may not yet initialized when wfx_tx_queues_get() is called.
--- drivers/staging/wfx/sta.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-)
drivers/staging/wfx/sta.c +14 -14
@@ -760,17 +760,6 @@ int wfx_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)return -EOPNOTSUPP ; } -for (i =0 ; i <ARRAY_SIZE (wdev->vif); i++) { -if (!wdev->vif[i]) { - wdev->vif[i] = vif; - wvif->id = i; -break ; - } - } -if (i ==ARRAY_SIZE (wdev->vif)) { -mutex_unlock (&wdev->conf_mutex); -return -EOPNOTSUPP ; - }// FIXME: prefer use of container_of() to get vif wvif->vif = vif; wvif->wdev = wdev;
@@ -787,12 +776,22 @@ int wfx_add_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif)init_completion (&wvif->scan_complete);INIT_WORK (&wvif->scan_work, wfx_hw_scan_work); -mutex_unlock (&wdev->conf_mutex); +wfx_tx_queues_init (wvif); +wfx_tx_policy_init (wvif); + +for (i =0 ; i <ARRAY_SIZE (wdev->vif); i++) { +if (!wdev->vif[i]) { + wdev->vif[i] = vif; + wvif->id = i; +break ; + } + } +WARN (i ==ARRAY_SIZE (wdev->vif),"try to instantiate more vif than supported" );hif_set_macaddr (wvif, vif->addr); -wfx_tx_queues_init (wvif); -wfx_tx_policy_init (wvif); +mutex_unlock (&wdev->conf_mutex); + wvif =NULL ;while ((wvif =wvif_iterate (wdev, wvif)) !=NULL ) {// Combo mode does not support Block Acks. We can re-enable them
@@ -824,6 +823,7 @@ void wfx_remove_interface(struct ieee80211_hw *hw, struct ieee80211_vif *vif) wvif->vif =NULL ;mutex_unlock (&wdev->conf_mutex); + wvif =NULL ;while ((wvif =wvif_iterate (wdev, wvif)) !=NULL ) {// Combo mode does not support Block Acks. We can re-enable them
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 46/147 media: platform: Improve queue set up flow for bug fixing
In reply to Sasha LevinFrom: Xia Jiang <[email protected]>
[ Upstream commit 5095a6413a0cf896ab468009b6142cb0fe617e66 ]
Add checking created buffer size follow in mtk_jpeg_queue_setup().
--- drivers/media/platform/mtk-jpeg/mtk_jpeg_core.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/media/platform/mtk-jpeg/mtk_jpeg_core.c +7 -0
@@ -571,6 +571,13 @@ static int mtk_jpeg_queue_setup(struct vb2_queue *q,if (!q_data)return -EINVAL ; +if (*num_planes) { +for (i =0 ; i < *num_planes; i++) +if (sizes[i] < q_data->sizeimage[i]) +return -EINVAL ; +return 0 ; + } + *num_planes = q_data->fmt->colplanes;for (i =0 ; i < q_data->fmt->colplanes; i++) { sizes[i] = q_data->sizeimage[i];
Signature · 7 lines
2.25.1 _______________________________________________ Linux-mediatek mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-mediatek
Sasha Levin
maintainerPATCH 47/147 usb: typec: tcpm: During PR_SWAP, source caps should be sent only after tSwapSourceStart
In reply to Sasha LevinFrom: Badhri Jagan Sridharan <[email protected]>
[ Upstream commit 6bbe2a90a0bb4af8dd99c3565e907fe9b5e7fd88 ]
The patch addresses the compliance test failures while running TD.PD.CP.E3, TD.PD.CP.E4, TD.PD.CP.E5 of the "Deterministic PD Compliance MOI" test plan published in https://www.usb.org/usbc.
For a product to be Type-C compliant, it's expected that these tests are run on usb.org certified Type-C compliance tester as mentioned in
https://www.usb.org/usbc.
The purpose of the tests TD.PD.CP.E3, TD.PD.CP.E4, TD.PD.CP.E5 is to verify the PR_SWAP response of the device. While doing so, the test asserts that Source Capabilities message is NOT received from the test device within tSwapSourceStart min (20 ms) from the time the last bit of GoodCRC corresponding to the RS_RDY message sent by the UUT was sent. If it does then the test fails.
This is in line with the requirements from the USB Power Delivery Specification Revision 3.0, Version 1.2:
"6.6.8.1 SwapSourceStartTimer
The SwapSourceStartTimer Shall be used by the new Source, after a Power Role Swap or Fast Role Swap, to ensure that it does not send Source_Capabilities Message before the new Sink is ready to receive theSource_Capabilities Message. The new Source Shall Not send the Source_Capabilities Message earlier than tSwapSourceStart after the last bit of the EOP of GoodCRC Message sent in response to the PS_RDY Message sent by the new Source indicating that its power supply is ready."
The patch makes sure that TCPM does not send the Source_Capabilities Message within tSwapSourceStart(20ms) by transitioning into SRC_STARTUP only after tSwapSourceStart(20ms).
--- drivers/usb/typec/tcpm/tcpm.c | 2 +- include/linux/usb/pd.h | 1 + 2 files changed, 2 insertions(+), 1 deletion(-)
drivers/usb/typec/tcpm/tcpm.c +1 -1
@@ -3573,7 +3573,7 @@ static void run_state_machine(struct tcpm_port *port)*/ tcpm_set_pwr_role (port,TYPEC_SOURCE );tcpm_pd_send_control (port,PD_CTRL_PS_RDY ); -tcpm_set_state (port,SRC_STARTUP ,0 ); +tcpm_set_state (port,SRC_STARTUP ,PD_T_SWAP_SRC_START );break ;case VCONN_SWAP_ACCEPT :
include/linux/usb/pd.h +1 -0
@@ -473,6 +473,7 @@ static inline unsigned int rdo_max_power(u32 rdo)#define PD_T_ERROR_RECOVERY 100/* minimum 25 is insufficient */ #define PD_T_SRCSWAPSTDBY 625/* Maximum of 650ms */ #define PD_T_NEWSRC 250/* Maximum of 275ms */ +#define PD_T_SWAP_SRC_START 20/* Minimum of 20ms */ #define PD_T_DRP_TRY 100/* 75 - 150 ms */ #define PD_T_DRP_TRYWAIT 600/* 400 - 800 ms */
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 48/147 media: tw5864: check status of tw5864_frameinterval_get
In reply to Sasha LevinFrom: Tom Rix <[email protected]>
[ Upstream commit 780d815dcc9b34d93ae69385a8465c38d423ff0f ]
clang static analysis reports this problem
tw5864-video.c:773:32: warning: The left expression of the compound
assignment is an uninitialized value.
The computed value will also be garbage
fintv->stepwise.max.numerator *= std_max_fps;
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ^stepwise.max is set with frameinterval, which comes from
ret = tw5864_frameinterval_get(input, &frameinterval); fintv->stepwise.step = frameinterval; fintv->stepwise.min = frameinterval; fintv->stepwise.max = frameinterval; fintv->stepwise.max.numerator *= std_max_fps;
When tw5864_frameinterval_get() fails, frameinterval is not set. So check the status and fix another similar problem.
--- drivers/media/pci/tw5864/tw5864-video.c | 6 ++++++ 1 file changed, 6 insertions(+)
drivers/media/pci/tw5864/tw5864-video.c +6 -0
@@ -767,6 +767,9 @@ static int tw5864_enum_frameintervals(struct file *file, void *priv, fintv->type =V4L2_FRMIVAL_TYPE_STEPWISE ; ret =tw5864_frameinterval_get (input, &frameinterval); +if (ret) +return ret; + fintv->stepwise.step = frameinterval; fintv->stepwise.min = frameinterval; fintv->stepwise.max = frameinterval;
@@ -785,6 +788,9 @@ static int tw5864_g_parm(struct file *file, void *priv, cp->capability =V4L2_CAP_TIMEPERFRAME ; ret =tw5864_frameinterval_get (input, &cp->timeperframe); +if (ret) +return ret; + cp->timeperframe.numerator *= input->frame_interval; cp->capturemode =0 ; cp->readbuffers =2 ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 49/147 drm/vkms: avoid warning in vkms_get_vblank_timestamp
In reply to Sasha LevinFrom: Sidong Yang <[email protected]>
[ Upstream commit 05ca530268a9d0ab3547e7b288635e35990a77c4 ]
This patch avoid the warning in vkms_get_vblank_timestamp when vblanks aren't enabled. When running igt test kms_cursor_crc just after vkms module, the warning raised like below. Initial value of vblank time is zero and hrtimer.node.expires is also zero if vblank aren't enabled before. vkms module isn't real hardware but just virtual hardware module. so vkms can't generate a resonable timestamp when hrtimer is off. it's best to grab the current time.
[106444.464503] [IGT] kms_cursor_crc: starting subtest pipe-A-cursor-size-change [106444.471475] WARNING: CPU: 0 PID: 10109 at vkms_get_vblank_timestamp+0x42/0x50 [vkms] [106444.471511] CPU: 0 PID: 10109 Comm: kms_cursor_crc Tainted: G W OE 5.9.0-rc1+ #6 [106444.471514] RIP: 0010:vkms_get_vblank_timestamp+0x42/0x50 [vkms] [106444.471528] Call Trace: [106444.471551] drm_get_last_vbltimestamp+0xb9/0xd0 [drm] [106444.471566] drm_reset_vblank_timestamp+0x63/0xe0 [drm] [106444.471579] drm_crtc_vblank_on+0x85/0x150 [drm] [106444.471582] vkms_crtc_atomic_enable+0xe/0x10 [vkms] [106444.471592] drm_atomic_helper_commit_modeset_enables+0x1db/0x230 [drm_kms_helper] [106444.471594] vkms_atomic_commit_tail+0x38/0xc0 [vkms] [106444.471601] commit_tail+0x97/0x130 [drm_kms_helper] [106444.471608] drm_atomic_helper_commit+0x117/0x140 [drm_kms_helper] [106444.471622] drm_atomic_commit+0x4a/0x50 [drm] [106444.471629] drm_atomic_helper_set_config+0x63/0xb0 [drm_kms_helper] [106444.471642] drm_mode_setcrtc+0x1d9/0x7b0 [drm] [106444.471654] ? drm_mode_getcrtc+0x1a0/0x1a0 [drm] [106444.471666] drm_ioctl_kernel+0xb6/0x100 [drm] [106444.471677] drm_ioctl+0x3ad/0x470 [drm] [106444.471688] ? drm_mode_getcrtc+0x1a0/0x1a0 [drm] [106444.471692] ? tomoyo_file_ioctl+0x19/0x20 [106444.471694] __x64_sys_ioctl+0x96/0xd0 [106444.471697] do_syscall_64+0x37/0x80 [106444.471699] entry_SYSCALL_64_after_hwframe+0x44/0xa9
--- drivers/gpu/drm/vkms/vkms_crtc.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/gpu/drm/vkms/vkms_crtc.c +5 -0
@@ -86,6 +86,11 @@ static bool vkms_get_vblank_timestamp(struct drm_crtc *crtc,struct vkms_output *output = &vkmsdev->output;struct drm_vblank_crtc *vblank = &dev->vblank[pipe]; +if (!READ_ONCE (vblank->enabled)) { + *vblank_time =ktime_get (); +return true; + } + *vblank_time =READ_ONCE (output->vblank_hrtimer.node.expires);if (WARN_ON (*vblank_time == vblank->time))
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 50/147 media: imx274: fix frame interval handling
In reply to Sasha LevinFrom: Hans Verkuil <[email protected]>
[ Upstream commit 49b20d981d723fae5a93843c617af2b2c23611ec ]
1) the numerator and/or denominator might be 0, in that case fall back to the default frame interval. This is per the spec and this caused a v4l2-compliance failure.
2) the updated frame interval wasn't returned in the s_frame_interval subdev op.
--- drivers/media/i2c/imx274.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/media/i2c/imx274.c +5 -3
@@ -1235,6 +1235,8 @@ static int imx274_s_frame_interval(struct v4l2_subdev *sd, ret =imx274_set_frame_interval (imx274, fi->interval);if (!ret) { + fi->interval = imx274->frame_interval; +/* * exposure time range is decided by frame interval * need to update it after frame interval changes
@@ -1730,9 +1732,9 @@ static int imx274_set_frame_interval(struct stimx274 *priv, __func__, frame_interval.numerator, frame_interval.denominator); -if (frame_interval.numerator ==0 ) { - err = -EINVAL ; - goto fail; +if (frame_interval.numerator ==0 || frame_interval.denominator ==0 ) { + frame_interval.denominator =IMX274_DEF_FRAME_RATE ; + frame_interval.numerator =1 ; } req_frame_rate = (u32)(frame_interval.denominator
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 51/147 mmc: via-sdmmc: Fix data race bug
In reply to Sasha LevinFrom: Madhuparna Bhowmik <[email protected]>
[ Upstream commit 87d7ad089b318b4f319bf57f1daa64eb6d1d10ad ]
via_save_pcictrlreg() should be called with host->lock held as it writes to pm_pcictrl_reg, otherwise there can be a race condition between via_sd_suspend() and via_sdc_card_detect(). The same pattern is used in the function via_reset_pcictrl() as well, where via_save_pcictrlreg() is called with host->lock held.
Found by Linux Driver Verification project (linuxtesting.org).
--- drivers/mmc/host/via-sdmmc.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/mmc/host/via-sdmmc.c +3 -0
@@ -1257,11 +1257,14 @@ static void __maybe_unused via_init_sdc_pm(struct via_crdr_mmc_host *host)static int __maybe_unused via_sd_suspend (struct device *dev) {struct via_crdr_mmc_host *host; +unsigned long flags; host =dev_get_drvdata (dev); +spin_lock_irqsave (&host->lock, flags);via_save_pcictrlreg (host);via_save_sdcreg (host); +spin_unlock_irqrestore (&host->lock, flags);device_wakeup_enable (dev);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 52/147 drm/bridge/synopsys: dsi: add support for non-continuous HS clock
In reply to Sasha LevinFrom: Antonio Borneo <[email protected]>
[ Upstream commit c6d94e37bdbb6dfe7e581e937a915ab58399b8a5 ]
Current code enables the HS clock when video mode is started or to send out a HS command, and disables the HS clock to send out a LP command. This is not what DSI spec specify.
Enable HS clock either in command and in video mode. Set automatic HS clock management for panels and devices that support non-continuous HS clock.
--- drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/gpu/drm/bridge/synopsys/dw-mipi-dsi.c +7 -2
@@ -365,7 +365,6 @@ static void dw_mipi_message_config(struct dw_mipi_dsi *dsi,if (lpm) val |=CMD_MODE_ALL_LP ; -dsi_write (dsi,DSI_LPCLK_CTRL , lpm ?0 :PHY_TXREQUESTCLKHS );dsi_write (dsi,DSI_CMD_MODE_CFG , val); }
@@ -541,16 +540,22 @@ static void dw_mipi_dsi_video_mode_config(struct dw_mipi_dsi *dsi)static void dw_mipi_dsi_set_mode (struct dw_mipi_dsi *dsi,unsigned long mode_flags) { +u32 val; +dsi_write (dsi,DSI_PWR_UP ,RESET );if (mode_flags &MIPI_DSI_MODE_VIDEO ) {dsi_write (dsi,DSI_MODE_CFG ,ENABLE_VIDEO_MODE );dw_mipi_dsi_video_mode_config (dsi); -dsi_write (dsi,DSI_LPCLK_CTRL ,PHY_TXREQUESTCLKHS ); }else {dsi_write (dsi,DSI_MODE_CFG ,ENABLE_CMD_MODE ); } + val =PHY_TXREQUESTCLKHS ; +if (dsi->mode_flags &MIPI_DSI_CLOCK_NON_CONTINUOUS ) + val |=AUTO_CLKLANE_CTRL ; +dsi_write (dsi,DSI_LPCLK_CTRL , val); +dsi_write (dsi,DSI_PWR_UP ,POWERUP ); }
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 53/147 brcmfmac: increase F2 watermark for BCM4329
In reply to Sasha LevinFrom: Dmitry Osipenko <[email protected]>
[ Upstream commit 317da69d10b0247c4042354eb90c75b81620ce9d ]
This patch fixes SDHCI CRC errors during of RX throughput testing on BCM4329 chip if SDIO BUS is clocked above 25MHz. In particular the checksum problem is observed on NVIDIA Tegra20 SoCs. The good watermark value is borrowed from downstream BCMDHD driver and it's matching to the value that is already used for the BCM4339 chip, hence let's re-use it for BCM4329.
--- drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 1 + 1 file changed, 1 insertion(+)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +1 -0
@@ -4278,6 +4278,7 @@ static void brcmf_sdio_firmware_callback(struct device *dev, int err,brcmf_sdiod_writeb (sdiod,SBSDIO_FUNC1_MESBUSYCTRL ,CY_43012_MESBUSYCTRL , &err);break ; +case SDIO_DEVICE_ID_BROADCOM_4329 :case SDIO_DEVICE_ID_BROADCOM_4339 :brcmf_dbg (INFO ,"set F2 watermark to 0x%x*4 bytes for 4339\n" ,CY_4339_F2_WATERMARK );
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 54/147 sfc: add and use efx_tx_send_pending in tx.c
In reply to Sasha LevinFrom: Edward Cree <[email protected]>
[ Upstream commit 1c0544d24927e4fad04f858216b8ea767a3bd123 ]
Instead of using efx_tx_queue_partner(), which relies on the assumption
that tx_queues_per_channel is 2, efx_tx_send_pending() iterates over txqs with efx_for_each_channel_tx_queue().
We unconditionally set tx_queue->xmit_pending (renamed from
xmit_more_available), then condition on xmit_more for the call to efx_tx_send_pending(), which will clear xmit_pending. Thus, after an xmit_more TX, the doorbell is un-rung and xmit_pending is true.
--- drivers/net/ethernet/sfc/ef10.c | 2 +- drivers/net/ethernet/sfc/ef100_tx.c | 14 +++---- drivers/net/ethernet/sfc/farch.c | 2 +- drivers/net/ethernet/sfc/net_driver.h | 4 +- drivers/net/ethernet/sfc/tx.c | 59 ++++++++++++++------------- drivers/net/ethernet/sfc/tx_common.c | 4 +- 6 files changed, 44 insertions(+), 41 deletions(-)
drivers/net/ethernet/sfc/ef10.c +1 -1
@@ -2367,7 +2367,7 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue)unsigned int write_ptr;efx_qword_t *txd; - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false;if (unlikely (tx_queue->write_count == tx_queue->insert_count))return ;
drivers/net/ethernet/sfc/ef100_tx.c +7 -7
@@ -131,7 +131,7 @@ void ef100_notify_tx_desc(struct efx_tx_queue *tx_queue)efx_writed_page (tx_queue->efx, ®,ER_GZ_TX_RING_DOORBELL , tx_queue->queue); tx_queue->notify_count = tx_queue->write_count; - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false; }static void ef100_tx_push_buffers (struct efx_tx_queue *tx_queue)
@@ -373,14 +373,14 @@ int ef100_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb) }if (__netdev_tx_sent_queue (tx_queue->core_txq, skb->len, xmit_more)) - tx_queue->xmit_more_available = false;/* push doorbell */ + tx_queue->xmit_pending = false;/* push doorbell */ else if (tx_queue->write_count - tx_queue->notify_count >255 )/* Ensure we never push more than 256 packets at once */ - tx_queue->xmit_more_available = false;/* push */ + tx_queue->xmit_pending = false;/* push */ else - tx_queue->xmit_more_available = true;/* don't push yet */ + tx_queue->xmit_pending = true;/* don't push yet */ -if (!tx_queue->xmit_more_available) +if (!tx_queue->xmit_pending)ef100_tx_push_buffers (tx_queue);if (segments) {
@@ -400,9 +400,9 @@ int ef100_enqueue_skb(struct efx_tx_queue *tx_queue, struct sk_buff *skb)/* If we're not expecting another transmit and we had something to push * on this queue then we need to push here to get the previous packets * out. We only enter this branch from before the 'Update BQL' section -* above, so xmit_more_available still refers to the old state. +* above, so xmit_pending still refers to the old state. */ -if (tx_queue->xmit_more_available && !xmit_more) +if (tx_queue->xmit_pending && !xmit_more)ef100_tx_push_buffers (tx_queue);return rc; }
drivers/net/ethernet/sfc/farch.c +1 -1
@@ -320,7 +320,7 @@ void efx_farch_tx_write(struct efx_tx_queue *tx_queue)unsigned write_ptr;unsigned old_write_count = tx_queue->write_count; - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false;if (unlikely (tx_queue->write_count == tx_queue->insert_count))return ;
drivers/net/ethernet/sfc/net_driver.h +2 -2
@@ -244,7 +244,7 @@ struct efx_tx_buffer {* @tso_fallbacks: Number of times TSO fallback used * @pushes: Number of times the TX push feature has been used * @pio_packets: Number of times the TX PIO feature has been used -* @xmit_more_available: Are any packets waiting to be pushed to the NIC +* @xmit_pending: Are any packets waiting to be pushed to the NIC * @cb_packets: Number of times the TX copybreak feature has been used * @notify_count: Count of notified descriptors to the NIC * @empty_read_count: If the completion path has seen the queue as empty
@@ -292,7 +292,7 @@ struct efx_tx_queue {unsigned int tso_fallbacks;unsigned int pushes;unsigned int pio_packets; -bool xmit_more_available; +bool xmit_pending;unsigned int cb_packets;unsigned int notify_count;/* Statistics to supplement MAC stats */
drivers/net/ethernet/sfc/tx.c +31 -28
90 lines not on this page · raw message
drivers/net/ethernet/sfc/tx_common.c +2 -2
@@ -78,7 +78,7 @@ void efx_init_tx_queue(struct efx_tx_queue *tx_queue) tx_queue->read_count =0 ; tx_queue->old_read_count =0 ; tx_queue->empty_read_count =0 |EFX_EMPTY_COUNT_VALID ; - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false; tx_queue->timestamping = (efx_ptp_use_mac_tx_timestamps (efx) && tx_queue->channel ==efx_ptp_channel (efx)); tx_queue->completed_timestamp_major =0 ;
@@ -116,7 +116,7 @@ void efx_fini_tx_queue(struct efx_tx_queue *tx_queue) ++tx_queue->read_count; } - tx_queue->xmit_more_available = false; + tx_queue->xmit_pending = false;netdev_tx_reset_queue (tx_queue->core_txq); }
Signature · 1 line
2.25.1
Re: [PATCH AUTOSEL 5.9 054/147] sfc: add and use efx_tx_send_pending in tx.c
Quoted 7 lines · Sasha Levin
From: Edward Cree <[email protected]>
[ Upstream commit
1c0544d24927e4fad04f858216b8ea767a3bd123]Instead of using
efx_tx_queue_partner(), which relies on the assumption
thattx_queues_per_channelis 2,efx_tx_send_pending()iterates over txqs withefx_for_each_channel_tx_queue().
That assumption was valid for the code as of v5.9; this change was only needed to support the extra queues that were added for encap offloads. Thus, this patch shouldn't be backported, unless -stable is also planning to backport that feature (e.g. 0ce8df661456, 24b2c3751aa3), which I doubt (it's nearly 20 patches, and can't be considered a bugfix).
-ed
Sasha Levin
maintainerRe: [PATCH AUTOSEL 5.9 054/147] sfc: add and use efx_tx_send_pending in tx.c
Quoted 5 lines · Edward Cree
Quoted 4 lines · Sasha Levin
From: Edward Cree <[email protected]>
[ Upstream commit
1c0544d24927e4fad04f858216b8ea767a3bd123]
Quoted 8 lines
Quoted 3 lines
Instead of using
efx_tx_queue_partner(), which relies on the assumption
thattx_queues_per_channelis 2,efx_tx_send_pending()iterates over txqs withefx_for_each_channel_tx_queue().That assumption was valid for the code as of v5.9; this change was only needed to support the extra queues that were added for encap offloads. Thus, this patch shouldn't be backported, unless -stable is also planning to backport that feature (e.g.
0ce8df661456,24b2c3751aa3), which I doubt (it's nearly 20 patches, and can't be considered a bugfix).
Right, now dropped - thanks!
Signature · 2 lines
Thanks, Sasha
Sasha Levin
maintainerPATCH 55/147 arm64: topology: Stop using MPIDR for topology information
In reply to Sasha LevinFrom: Valentin Schneider <[email protected]>
[ Upstream commit 3102bc0e6ac752cc5df896acb557d779af4d82a1 ]
In the absence of ACPI or DT topology data, we fallback to haphazardly decoding *something* out of MPIDR. Sadly, the contents of that register are mostly unusable due to the implementation leniancy and things like Aff0 having to be capped to 15 (despite being encoded on 8 bits).
Consider a simple system with a single package of 32 cores, all under the same LLC. We ought to be shoving them in the same core_sibling mask, but MPIDR is going to look like:
| CPU | 0 | ... | 15 | 16 | ... | 31 | |------+---+-----+----+----+-----+----+ | Aff0 | 0 | ... | 15 | 0 | ... | 15 | | Aff1 | 0 | ... | 0 | 1 | ... | 1 | | Aff2 | 0 | ... | 0 | 0 | ... | 0 |
Which will eventually yield
core_sibling(0-15) == 0-15
core_sibling(16-31) == 16-31
NUMA woes
=========
If we try to play games with this and set up NUMA boundaries within those groups of 16 cores via e.g. QEMU:
# Node0: 0-9; Node1: 10-19
$ qemu-system-aarch64 <blah> \
-smp 20 -numa node,cpus=0-9,nodeid=0 -numa node,cpus=10-19,nodeid=1The scheduler's MC domain (all CPUs with same LLC) is going to be built via
arch_topology.c::cpu_coregroup_mask()
In there we try to figure out a sensible mask out of the topology information we have. In short, here we'll pick the smallest of NUMA or core sibling mask.
node_mask(CPU9) == 0-9
core_sibling(CPU9) == 0-15
MC mask for CPU9 will thus be 0-9, not a problem.
node_mask(CPU10) == 10-19
core_sibling(CPU10) == 0-15
MC mask for CPU10 will thus be 10-19, not a problem.
node_mask(CPU16) == 10-19
core_sibling(CPU16) == 16-19
MC mask for CPU16 will thus be 16-19... Uh oh. CPUs 16-19 are in two different unique MC spans, and the scheduler has no idea what to make of that. That triggers the WARN_ON() added by commit
ccf74128d66c ("sched/topology: Assert non-NUMA topology masks don't (partially) overlap")
Fixing MPIDR-derived topology
=============================
We could try to come up with some cleverer scheme to figure out which of the available masks to pick, but really if one of those masks resulted from MPIDR then it should be discarded because it's bound to be bogus.
I was hoping to give MPIDR a chance for SMT, to figure out which threads are in the same core using Aff1-3 as core ID, but Sudeep and Robin pointed out to me that there are systems out there where *all* cores have non-zero values in their higher affinity fields (e.g. RK3288 has "5" in all of its cores' MPIDR.Aff1), which would expose a bogus core ID to userspace.
Stop using MPIDR for topology information. When no other source of topology information is available, mark each CPU as its own core and its NUMA node as its LLC domain.
--- arch/arm64/kernel/topology.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-)
arch/arm64/kernel/topology.c +17 -15
@@ -36,21 +36,23 @@ void store_cpu_topology(unsigned int cpuid)if (mpidr &MPIDR_UP_BITMASK )return ; -/* Create cpu topology mapping based on MPIDR. */ -if (mpidr &MPIDR_MT_BITMASK ) { -/* Multiprocessor system : Multi-threads per core */ - cpuid_topo->thread_id =MPIDR_AFFINITY_LEVEL (mpidr,0 ); - cpuid_topo->core_id =MPIDR_AFFINITY_LEVEL (mpidr,1 ); - cpuid_topo->package_id =MPIDR_AFFINITY_LEVEL (mpidr,2 ) | -MPIDR_AFFINITY_LEVEL (mpidr,3 ) <<8 ; - }else { -/* Multiprocessor system : Single-thread per core */ - cpuid_topo->thread_id =-1 ; - cpuid_topo->core_id =MPIDR_AFFINITY_LEVEL (mpidr,0 ); - cpuid_topo->package_id =MPIDR_AFFINITY_LEVEL (mpidr,1 ) | -MPIDR_AFFINITY_LEVEL (mpidr,2 ) <<8 | -MPIDR_AFFINITY_LEVEL (mpidr,3 ) <<16 ; - } +/* +* This would be the place to create cpu topology based on MPIDR. +* +* However, it cannot be trusted to depict the actual topology; some +* pieces of the architecture enforce an artificial cap on Aff0 values +* (e.g. GICv3's ICC_SGI1R_EL1 limits it to 15), leading to an +* artificial cycling of Aff1, Aff2 and Aff3 values. IOW, these end up +* having absolutely no relationship to the actual underlying system +* topology, and cannot be reasonably used as core / package ID. +* +* If the MT bit is set, Aff0 *could* be used to define a thread ID, but +* we still wouldn't be able to obtain a sane core ID. This means we +* need to entirely ignore MPIDR for any topology deduction. +*/ + cpuid_topo->thread_id =-1 ; + cpuid_topo->core_id = cpuid; + cpuid_topo->package_id =cpu_to_node (cpuid); pr_debug("CPU%u: cluster %d core %d thread %d mpidr %#016llx\n", cpuid, cpuid_topo->package_id, cpuid_topo->core_id,
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 56/147 printk: reduce LOG_BUF_SHIFT range for H8300
In reply to Sasha LevinFrom: John Ogness <[email protected]>
[ Upstream commit 550c10d28d21bd82a8bb48debbb27e6ed53262f6 ]
The .bss section for the h8300 is relatively small. A value of CONFIG_LOG_BUF_SHIFT that is larger than 19 will create a static printk ringbuffer that is too large. Limit the range appropriately for the H8300.
--- init/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
init/Kconfig +2 -1
@@ -682,7 +682,8 @@ config IKHEADERS config LOG_BUF_SHIFT int "Kernel log buffer size (16 => 64KB, 17 => 128KB)" - range 12 25 + range 12 25 if !H8300 + range 12 19 if H8300 default 17 depends on PRINTK help
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 57/147 ia64: kprobes: Use generic kretprobe trampoline handler
In reply to Sasha LevinFrom: Masami Hiramatsu <[email protected]>
[ Upstream commit e792ff804f49720ce003b3e4c618b5d996256a18 ]
Use the generic kretprobe trampoline handler. Don't use framepointer verification.
--- arch/ia64/kernel/kprobes.c | 77 +------------------------------------- 1 file changed, 2 insertions(+), 75 deletions(-)
arch/ia64/kernel/kprobes.c +2 -74
@@ -396,83 +396,9 @@ static void kretprobe_trampoline(void) { } -/* -* At this point the target function has been tricked into -* returning into our trampoline. Lookup the associated instance -* and then: -* - call the handler function -* - cleanup by marking the instance as unused -* - long jump back to the original return address -*/ int __kprobes trampoline_probe_handler (struct kprobe *p,struct pt_regs *regs) { -struct kretprobe_instance *ri =NULL ; -struct hlist_head *head, empty_rp; -struct hlist_node *tmp; -unsigned long flags, orig_ret_address =0 ; - unsigned long trampoline_address - ((struct fnptr *)kretprobe_trampoline)->ip; - -INIT_HLIST_HEAD (&empty_rp); -kretprobe_hash_lock (current, &head, &flags); - -/* -* It is possible to have multiple instances associated with a given -* task either because an multiple functions in the call path -* have a return probe installed on them, and/or more than one return -* return probe was registered for a target function. -* -* We can handle this because: -* - instances are always inserted at the head of the list -* - when multiple return probes are registered for the same -* function, the first instance's ret_addr will point to the -* real return address, and all the rest will point to -* kretprobe_trampoline -*/ -hlist_for_each_entry_safe (ri, tmp, head, hlist) { -if (ri->task != current) -/* another task is sharing our hash bucket */ -continue ; - - orig_ret_address = (unsigned long )ri->ret_addr; -if (orig_ret_address != trampoline_address) -/* -* This is the real return address. Any other -* instances associated with this task are for -* other calls deeper on the call stack -*/ -break ; - } - - regs->cr_iip = orig_ret_address; - -hlist_for_each_entry_safe (ri, tmp, head, hlist) { -if (ri->task != current) -/* another task is sharing our hash bucket */ -continue ; - -if (ri->rp && ri->rp->handler) - ri->rp->handler (ri, regs); - - orig_ret_address = (unsigned long )ri->ret_addr; -recycle_rp_inst (ri, &empty_rp); - -if (orig_ret_address != trampoline_address) -/* -* This is the real return address. Any other -* instances associated with this task are for -* other calls deeper on the call stack -*/ -break ; - } -kretprobe_assert (ri, orig_ret_address, trampoline_address); - -kretprobe_hash_unlock (current, &flags); - -hlist_for_each_entry_safe (ri, tmp, &empty_rp, hlist) { -hlist_del (&ri->hlist); -kfree (ri); - } + regs->cr_iip =__kretprobe_trampoline_handler (regs, kretprobe_trampoline,NULL );/* * By returning a non-zero value, we are telling * kprobe_handler() that we don't want the post_handler
@@ -485,6 +411,7 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, struct pt_regs *regs) { ri->ret_addr = (kprobe_opcode_t *)regs->b0; + ri->fp =NULL ;/* Replace the return addr with trampoline addr */ regs->b0 = ((struct fnptr *)kretprobe_trampoline)->ip;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 58/147 selftests/powerpc: Make using_hash_mmu() work on Cell & PowerMac
In reply to Sasha LevinFrom: Michael Ellerman <[email protected]>
[ Upstream commit 34c103342be3f9397e656da7c5cc86e97b91f514 ]
These platforms don't show the MMU in /proc/cpuinfo, but they always use hash, so teach using_hash_mmu() that.
--- tools/testing/selftests/powerpc/utils.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
tools/testing/selftests/powerpc/utils.c +3 -1
@@ -318,7 +318,9 @@ int using_hash_mmu(bool *using_hash) rc =0 ;while (fgets (line,sizeof (line), f) !=NULL ) { -if (strcmp (line,"MMU : Hash\n" ) ==0 ) { +if (!strcmp (line,"MMU : Hash\n" ) || + !strcmp (line,"platform : Cell\n" ) || + !strcmp (line,"platform : PowerMac\n" )) { *using_hash = true; goto out; }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 59/147 kgdb: Make "kgdbcon" work properly with "kgdb_earlycon"
In reply to Sasha LevinFrom: Douglas Anderson <[email protected]>
[ Upstream commit b18b099e04f450cdc77bec72acefcde7042bd1f3 ]
On my system the kernel processes the "kgdb_earlycon" parameter before the "kgdbcon" parameter. When we setup "kgdb_earlycon" we'll end up in kgdb_register_callbacks() and "kgdb_use_con" won't have been set yet so we'll never get around to starting "kgdbcon". Let's remedy this by detecting that the IO module was already registered when setting "kgdb_use_con" and registering the console then.
As part of this, to avoid pre-declaring things, move the handling of the "kgdbcon" further down in the file.
--- kernel/debug/debug_core.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-)
kernel/debug/debug_core.c +14 -8
@@ -94,14 +94,6 @@ int dbg_switch_cpu;/* Use kdb or gdbserver mode */ int dbg_kdb_mode =1 ; -static int __init opt_kgdb_con (char *str) -{ - kgdb_use_con =1 ; -return 0 ; -} - -early_param ("kgdbcon" , opt_kgdb_con); -module_param (kgdb_use_con, int,0644 );module_param (kgdbreboot, int,0644 );
@@ -920,6 +912,20 @@ static struct console kgdbcons = { .index = -1, }; +static int __init opt_kgdb_con (char *str) +{ + kgdb_use_con =1 ; + +if (kgdb_io_module_registered && !kgdb_con_registered) { +register_console (&kgdbcons); + kgdb_con_registered =1 ; + } + +return 0 ; +} + +early_param ("kgdbcon" , opt_kgdb_con); +#ifdef CONFIG_MAGIC_SYSRQ static void sysrq_handle_dbg (int key) {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 60/147 bpf: Permit map_ptr arithmetic with opcode add and offset 0
In reply to Sasha LevinFrom: Yonghong Song <[email protected]>
[ Upstream commit 7c6967326267bd5c0dded0a99541357d70dd11ac ]
Commit 41c48f3a98231 ("bpf: Support access
to bpf map fields") added support to access map fields with CORE support. For example,
struct bpf_map {
__u32 max_entries;
} __attribute__((preserve_access_index));
struct bpf_array {
struct bpf_map map;
__u32 elem_size;
} __attribute__((preserve_access_index));
struct {
__uint(type, BPF_MAP_TYPE_ARRAY);
__uint(max_entries, 4);
__type(key, __u32);
__type(value, __u32);
} m_array SEC(".maps");
SEC("cgroup_skb/egress")
int cg_skb(void *ctx)
{
struct bpf_array *array = (struct bpf_array *)&m_array;
/* .. array->map.max_entries .. */
}In kernel, bpf_htab has similar structure,
struct bpf_htab {
struct bpf_map map;
...
}In the above cg_skb(), to access array->map.max_entries, with CORE, the clang will generate two builtin's.
base = &m_array;
/* access array.map */
map_addr = __builtin_preserve_struct_access_info(base, 0, 0);
/* access array.map.max_entries */
max_entries_addr = __builtin_preserve_struct_access_info(map_addr, 0, 0);
max_entries = *max_entries_addr;In the current llvm, if two builtin's are in the same function or
in the same function after inlining, the compiler is smart enough to chain them together and generates like below:
base = &m_array;
max_entries = *(base + reloc_offset); /* reloc_offset = 0 in this case */
and we are fine.
But if we force no inlining for one of functions in test_map_ptr() selftest, e.g., check_default(), the above two __builtin_preserve_* will be in two different functions. In this case, we will have code like:
func check_hash():
reloc_offset_map = 0;
base = &m_array;
map_base = base + reloc_offset_map;
check_default(map_base, ...)
func check_default(map_base, ...):
max_entries = *(map_base + reloc_offset_max_entries);In kernel, map_ptr (CONST_PTR_TO_MAP) does not allow any arithmetic.
The above "map_base = base + reloc_offset_map" will trigger a verifier failure.
; VERIFY(check_default(&hash->map, map));
0: (18) r7 = 0xffffb4fe8018a004
2: (b4) w1 = 110
3: (63) *(u32 *)(r7 +0) = r1
R1_w=invP110 R7_w=map_value(id=0,off=4,ks=4,vs=8,imm=0) R10=fp0
; VERIFY_TYPE(BPF_MAP_TYPE_HASH, check_hash);
4: (18) r1 = 0xffffb4fe8018a000
6: (b4) w2 = 1
7: (63) *(u32 *)(r1 +0) = r2
R1_w=map_value(id=0,off=0,ks=4,vs=8,imm=0) R2_w=invP1 R7_w=map_value(id=0,off=4,ks=4,vs=8,imm=0) R10=fp0
8: (b7) r2 = 0
9: (18) r8 = 0xffff90bcb500c000
11: (18) r1 = 0xffff90bcb500c000
13: (0f) r1 += r2
R1 pointer arithmetic on map_ptr prohibited
To fix the issue, let us permit map_ptr + 0 arithmetic which will result in exactly the same map_ptr.
--- kernel/bpf/verifier.c | 4 ++++ 1 file changed, 4 insertions(+)
kernel/bpf/verifier.c +4 -0
@@ -5219,6 +5219,10 @@ static int adjust_ptr_min_max_vals(struct bpf_verifier_env *env, dst, reg_type_str[ptr_reg->type]);return -EACCES ;case CONST_PTR_TO_MAP : +/* smin_val represents the known value */ +if (known && smin_val ==0 && opcode ==BPF_ADD ) +break ; +/* fall-through */ case PTR_TO_PACKET_END :case PTR_TO_SOCKET :case PTR_TO_SOCKET_OR_NULL :
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 61/147 drm: exynos: fix common struct sg_table related issues
In reply to Sasha LevinFrom: Marek Szyprowski <[email protected]>
[ Upstream commit 84404614167b829f7b58189cd24b6c0c74897171 ]
The Documentation/DMA-API-HOWTO.txt states that the dma_map_sg() function returns the number of the created entries in the DMA address space. However the subsequent calls to the dma_sync_sg_for_{device,cpu}() and dma_unmap_sg must be called with the original number of the entries passed to the dma_map_sg().
struct sg_table is a common structure used for describing a non-contiguous memory buffer, used commonly in the DRM and graphics subsystems. It consists of a scatterlist with memory pages and DMA addresses (sgl entry), as well as the number of scatterlist entries: CPU pages (orig_nents entry) and DMA mapped pages (nents entry).
It turned out that it was a common mistake to misuse nents and orig_nents entries, calling DMA-mapping functions with a wrong number of entries or ignoring the number of mapped entries returned by the dma_map_sg() function.
To avoid such issues, lets use a common dma-mapping wrappers operating directly on the struct sg_table objects and use scatterlist page iterators where possible. This, almost always, hides references to the nents and orig_nents entries, making the code robust, easier to follow and copy/paste safe.
--- drivers/gpu/drm/exynos/exynos_drm_g2d.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/gpu/drm/exynos/exynos_drm_g2d.c +5 -5
@@ -395,8 +395,8 @@ static void g2d_userptr_put_dma_addr(struct g2d_data *g2d,return ; out: -dma_unmap_sg (to_dma_dev (g2d->drm_dev), g2d_userptr->sgt->sgl, - g2d_userptr->sgt->nents,DMA_BIDIRECTIONAL ); +dma_unmap_sgtable (to_dma_dev (g2d->drm_dev), g2d_userptr->sgt, +DMA_BIDIRECTIONAL ,0 ); pages =frame_vector_pages (g2d_userptr->vec);if (!IS_ERR (pages)) {
@@ -511,10 +511,10 @@ static dma_addr_t *g2d_userptr_get_dma_addr(struct g2d_data *g2d, g2d_userptr->sgt = sgt; -if (!dma_map_sg (to_dma_dev (g2d->drm_dev), sgt->sgl, sgt->nents, -DMA_BIDIRECTIONAL )) { + ret =dma_map_sgtable (to_dma_dev (g2d->drm_dev), sgt, +DMA_BIDIRECTIONAL ,0 ); +if (ret) {DRM_DEV_ERROR (g2d->dev,"failed to map sgt with dma region.\n" ); - ret = -ENOMEM ; goto err_sg_free_table; }
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 62/147 xen: gntdev: fix common struct sg_table related issues
In reply to Sasha LevinFrom: Marek Szyprowski <[email protected]>
[ Upstream commit d1749eb1ab85e04e58c29e58900e3abebbdd6e82 ]
The Documentation/DMA-API-HOWTO.txt states that the dma_map_sg() function returns the number of the created entries in the DMA address space. However the subsequent calls to the dma_sync_sg_for_{device,cpu}() and dma_unmap_sg must be called with the original number of the entries passed to the dma_map_sg().
struct sg_table is a common structure used for describing a non-contiguous memory buffer, used commonly in the DRM and graphics subsystems. It consists of a scatterlist with memory pages and DMA addresses (sgl entry), as well as the number of scatterlist entries: CPU pages (orig_nents entry) and DMA mapped pages (nents entry).
It turned out that it was a common mistake to misuse nents and orig_nents entries, calling DMA-mapping functions with a wrong number of entries or ignoring the number of mapped entries returned by the dma_map_sg() function.
To avoid such issues, lets use a common dma-mapping wrappers operating directly on the struct sg_table objects and use scatterlist page iterators where possible. This, almost always, hides references to the nents and orig_nents entries, making the code robust, easier to follow and copy/paste safe.
--- drivers/xen/gntdev-dmabuf.c | 13 ++++++------- 1 file changed, 6 insertions(+), 7 deletions(-)
drivers/xen/gntdev-dmabuf.c +6 -7
@@ -247,10 +247,9 @@ static void dmabuf_exp_ops_detach(struct dma_buf *dma_buf,if (sgt) {if (gntdev_dmabuf_attach->dir !=DMA_NONE ) -dma_unmap_sg_attrs (attach->dev, sgt->sgl, - sgt->nents, - gntdev_dmabuf_attach->dir, -DMA_ATTR_SKIP_CPU_SYNC ); +dma_unmap_sgtable (attach->dev, sgt, + gntdev_dmabuf_attach->dir, +DMA_ATTR_SKIP_CPU_SYNC );sg_free_table (sgt); }
@@ -288,8 +287,8 @@ dmabuf_exp_ops_map_dma_buf(struct dma_buf_attachment *attach, sgt =dmabuf_pages_to_sgt (gntdev_dmabuf->pages, gntdev_dmabuf->nr_pages);if (!IS_ERR (sgt)) { -if (!dma_map_sg_attrs (attach->dev, sgt->sgl, sgt->nents, dir, -DMA_ATTR_SKIP_CPU_SYNC )) { +if (dma_map_sgtable (attach->dev, sgt, dir, +DMA_ATTR_SKIP_CPU_SYNC )) {sg_free_table (sgt);kfree (sgt); sgt =ERR_PTR (-ENOMEM );
@@ -633,7 +632,7 @@ dmabuf_imp_to_refs(struct gntdev_dmabuf_priv *priv, struct device *dev,/* Now convert sgt to array of pages and check for page validity. */ i =0 ; -for_each_sg_page (sgt->sgl, &sg_iter, sgt->nents,0 ) { +for_each_sgtable_page (sgt, &sg_iter,0 ) {struct page *page =sg_page_iter_page (&sg_iter);/* * Check if page is valid: this can happen if we are given
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 63/147 drm: lima: fix common struct sg_table related issues
In reply to Sasha LevinFrom: Marek Szyprowski <[email protected]>
[ Upstream commit c3d9c17f486d5c54940487dc31a54ebfdeeb371a ]
The Documentation/DMA-API-HOWTO.txt states that the dma_map_sg() function returns the number of the created entries in the DMA address space. However the subsequent calls to the dma_sync_sg_for_{device,cpu}() and dma_unmap_sg must be called with the original number of the entries passed to the dma_map_sg().
struct sg_table is a common structure used for describing a non-contiguous memory buffer, used commonly in the DRM and graphics subsystems. It consists of a scatterlist with memory pages and DMA addresses (sgl entry), as well as the number of scatterlist entries: CPU pages (orig_nents entry) and DMA mapped pages (nents entry).
It turned out that it was a common mistake to misuse nents and orig_nents entries, calling DMA-mapping functions with a wrong number of entries or ignoring the number of mapped entries returned by the dma_map_sg() function.
To avoid such issues, lets use a common dma-mapping wrappers operating directly on the struct sg_table objects and use scatterlist page iterators where possible. This, almost always, hides references to the nents and orig_nents entries, making the code robust, easier to follow and copy/paste safe.
--- drivers/gpu/drm/lima/lima_gem.c | 11 ++++++++--- drivers/gpu/drm/lima/lima_vm.c | 5 ++--- 2 files changed, 10 insertions(+), 6 deletions(-)
drivers/gpu/drm/lima/lima_gem.c +8 -3
@@ -69,8 +69,7 @@ int lima_heap_alloc(struct lima_bo *bo, struct lima_vm *vm)return ret;if (bo->base.sgt) { -dma_unmap_sg (dev, bo->base.sgt->sgl, - bo->base.sgt->nents,DMA_BIDIRECTIONAL ); +dma_unmap_sgtable (dev, bo->base.sgt,DMA_BIDIRECTIONAL ,0 );sg_free_table (bo->base.sgt); }else { bo->base.sgt =kmalloc (sizeof (*bo->base.sgt),GFP_KERNEL );
@@ -80,7 +79,13 @@ int lima_heap_alloc(struct lima_bo *bo, struct lima_vm *vm) } } -dma_map_sg (dev, sgt.sgl, sgt.nents,DMA_BIDIRECTIONAL ); + ret =dma_map_sgtable (dev, &sgt,DMA_BIDIRECTIONAL ,0 ); +if (ret) { +sg_free_table (&sgt); +kfree (bo->base.sgt); + bo->base.sgt =NULL ; +return ret; + } *bo->base.sgt = sgt;
drivers/gpu/drm/lima/lima_vm.c +2 -3
@@ -124,7 +124,7 @@ int lima_vm_bo_add(struct lima_vm *vm, struct lima_bo *bo, bool create)if (err) goto err_out1; -for_each_sg_dma_page (bo->base.sgt->sgl, &sg_iter, bo->base.sgt->nents,0 ) { +for_each_sgtable_dma_page (bo->base.sgt, &sg_iter,0 ) { err =lima_vm_map_page (vm,sg_page_iter_dma_address (&sg_iter), bo_va->node.start + offset);if (err)
@@ -298,8 +298,7 @@ int lima_vm_map_bo(struct lima_vm *vm, struct lima_bo *bo, int pageoff)mutex_lock (&vm->lock); base = bo_va->node.start + (pageoff <<PAGE_SHIFT ); -for_each_sg_dma_page (bo->base.sgt->sgl, &sg_iter, - bo->base.sgt->nents, pageoff) { +for_each_sgtable_dma_page (bo->base.sgt, &sg_iter, pageoff) { err =lima_vm_map_page (vm,sg_page_iter_dma_address (&sg_iter), base + offset);if (err)
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 64/147 drm: panfrost: fix common struct sg_table related issues
In reply to Sasha LevinFrom: Marek Szyprowski <[email protected]>
[ Upstream commit 34a4e66faf8b22c8409cbd46839ba5e488b1e6a9 ]
The Documentation/DMA-API-HOWTO.txt states that the dma_map_sg() function returns the number of the created entries in the DMA address space. However the subsequent calls to the dma_sync_sg_for_{device,cpu}() and dma_unmap_sg must be called with the original number of the entries passed to the dma_map_sg().
struct sg_table is a common structure used for describing a non-contiguous memory buffer, used commonly in the DRM and graphics subsystems. It consists of a scatterlist with memory pages and DMA addresses (sgl entry), as well as the number of scatterlist entries: CPU pages (orig_nents entry) and DMA mapped pages (nents entry).
It turned out that it was a common mistake to misuse nents and orig_nents entries, calling DMA-mapping functions with a wrong number of entries or ignoring the number of mapped entries returned by the dma_map_sg() function.
To avoid such issues, lets use a common dma-mapping wrappers operating directly on the struct sg_table objects and use scatterlist page iterators where possible. This, almost always, hides references to the nents and orig_nents entries, making the code robust, easier to follow and copy/paste safe.
--- drivers/gpu/drm/panfrost/panfrost_gem.c | 4 ++-- drivers/gpu/drm/panfrost/panfrost_mmu.c | 7 +++---- 2 files changed, 5 insertions(+), 6 deletions(-)
drivers/gpu/drm/panfrost/panfrost_gem.c +2 -2
@@ -41,8 +41,8 @@ static void panfrost_gem_free_object(struct drm_gem_object *obj)for (i =0 ; i < n_sgt; i++) {if (bo->sgts[i].sgl) { -dma_unmap_sg (pfdev->dev, bo->sgts[i].sgl, - bo->sgts[i].nents,DMA_BIDIRECTIONAL ); +dma_unmap_sgtable (pfdev->dev, &bo->sgts[i], +DMA_BIDIRECTIONAL ,0 );sg_free_table (&bo->sgts[i]); } }
drivers/gpu/drm/panfrost/panfrost_mmu.c +3 -4
@@ -253,7 +253,7 @@ static int mmu_map_sg(struct panfrost_device *pfdev, struct panfrost_mmu *mmu,struct io_pgtable_ops *ops = mmu->pgtbl_ops;u64 start_iova = iova; -for_each_sg (sgt->sgl, sgl, sgt->nents, count) { +for_each_sgtable_dma_sg (sgt, sgl, count) {unsigned long paddr =sg_dma_address (sgl);size_t len =sg_dma_len (sgl);
@@ -517,10 +517,9 @@ static int panfrost_mmu_map_fault_addr(struct panfrost_device *pfdev, int as,if (ret) goto err_pages; -if (!dma_map_sg (pfdev->dev, sgt->sgl, sgt->nents,DMA_BIDIRECTIONAL )) { - ret = -EINVAL ; + ret =dma_map_sgtable (pfdev->dev, sgt,DMA_BIDIRECTIONAL ,0 ); +if (ret) goto err_map; - }mmu_map_sg (pfdev, bomapping->mmu, addr,IOMMU_WRITE |IOMMU_READ |IOMMU_NOEXEC , sgt);
Signature · 6 lines
2.25.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 65/147 media: uvcvideo: Fix dereference of out-of-bound list iterator
In reply to Sasha LevinFrom: "Daniel W. S. Almeida" <[email protected]>
[ Upstream commit f875bcc375c738bf2f599ff2e1c5b918dbd07c45 ]
Fixes the following coccinelle report:
drivers/media/usb/uvc/uvc_ctrl.c:1860:5-11:
ERROR: invalid reference to the index variable of the iterator on line 1854
by adding a boolean variable to check if the loop has found the
Found using - Coccinelle (http://coccinelle.lip6.fr)
[Replace cursor variable with bool found]
--- drivers/media/usb/uvc/uvc_ctrl.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/media/usb/uvc/uvc_ctrl.c +9 -4
@@ -1844,30 +1844,35 @@ int uvc_xu_ctrl_query(struct uvc_video_chain *chain, {struct uvc_entity *entity;struct uvc_control *ctrl; -unsigned int i, found =0 ; +unsigned int i; +bool found;u32 reqflags;u16 size;u8 *data =NULL ;int ret;/* Find the extension unit. */ + found = false;list_for_each_entry (entity, &chain->entities, chain) {if (UVC_ENTITY_TYPE (entity) ==UVC_VC_EXTENSION_UNIT && - entity->id == xqry->unit) + entity->id == xqry->unit) { + found = true;break ; + } } -if (entity->id != xqry->unit) { +if (!found) {uvc_trace (UVC_TRACE_CONTROL ,"Extension unit %u not found.\n" , xqry->unit);return -ENOENT ; }/* Find the control and perform delayed initialization if needed. */ + found = false;for (i =0 ; i < entity->ncontrols; ++i) { ctrl = &entity->controls[i];if (ctrl->index == xqry->selector -1 ) { - found =1 ; + found = true;break ; } }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 66/147 nfc: s3fwrn5: Add missing CRYPTO_HASH dependency
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 4aa62c62d4c41d71b2bda5ed01b78961829ee93c ]
The driver uses crypto hash functions so it needs to select CRYPTO_HASH. This fixes build errors:
arc-linux-ld: drivers/nfc/s3fwrn5/firmware.o: in function `s3fwrn5_fw_download':
firmware.c:(.text+0x152): undefined reference to `crypto_alloc_shash'
--- drivers/nfc/s3fwrn5/Kconfig | 1 + 1 file changed, 1 insertion(+)
drivers/nfc/s3fwrn5/Kconfig +1 -0
@@ -2,6 +2,7 @@ config NFC_S3FWRN5 tristate select CRYPTO + select CRYPTO_HASH help Core driver for Samsung S3FWRN5 NFC chip. Contains core utilities of chip. It's intended to be used by PHYs to avoid duplicating lots
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 67/147 selftests/bpf: Define string const as global for test_sysctl_prog.c
In reply to Sasha LevinFrom: Yonghong Song <[email protected]>
[ Upstream commit 6e057fc15a2da4ee03eb1fa6889cf687e690106e ]
When tweaking llvm optimizations, I found that selftest build failed
with the following error:
libbpf: elf: skipping unrecognized data section(6) .rodata.str1.1
libbpf: prog 'sysctl_tcp_mem': bad map relo against '.L__const.is_tcp_mem.tcp_mem_name'
in section '.rodata.str1.1'
Error: failed to open BPF object file: Relocation failed
make: *** [/work/net-next/tools/testing/selftests/bpf/test_sysctl_prog.skel.h] Error 255
make: *** Deleting file `/work/net-next/tools/testing/selftests/bpf/test_sysctl_prog.skel.h'
The local string constant "tcp_mem_name" is put into '.rodata.str1.1' section which libbpf cannot handle. Using untweaked upstream llvm, "tcp_mem_name" is completely inlined after loop unrolling.
Commit 7fb5eefd7639 ("selftests/bpf: Fix test_sysctl_loop{1, 2} failure due to clang change") solved a similar problem by defining the string const as a global. Let us do the same here
for test_sysctl_prog.c so it can weather future potential llvm changes.
--- tools/testing/selftests/bpf/progs/test_sysctl_prog.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/testing/selftests/bpf/progs/test_sysctl_prog.c +2 -2
@@ -19,11 +19,11 @@#define ARRAY_SIZE (x) (sizeof(x) / sizeof((x)[0]))#endif +const char tcp_mem_name[] ="net/ipv4/tcp_mem" ; static __always_inline int is_tcp_mem(struct bpf_sysctl *ctx) { -char tcp_mem_name[] ="net/ipv4/tcp_mem" ;unsigned char i; -char name[64 ]; +char name[sizeof (tcp_mem_name)];int ret;memset (name,0 ,sizeof (name));
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 68/147 selinux: access policycaps with READ_ONCE/WRITE_ONCE
In reply to Sasha LevinFrom: Stephen Smalley <[email protected]>
[ Upstream commit e8ba53d0023a76ba0f50e6ee3e6288c5442f9d33 ]
Use READ_ONCE/WRITE_ONCE for all accesses to the selinux_state.policycaps booleans to prevent compiler mischief.
--- security/selinux/include/security.h | 14 +++++++------- security/selinux/ss/services.c | 3 ++- 2 files changed, 9 insertions(+), 8 deletions(-)
security/selinux/include/security.h +7 -7
@@ -177,49 +177,49 @@ static inline bool selinux_policycap_netpeer(void) {struct selinux_state *state = &selinux_state; -return state->policycap[POLICYDB_CAPABILITY_NETPEER ]; +return READ_ONCE (state->policycap[POLICYDB_CAPABILITY_NETPEER ]); }static inline bool selinux_policycap_openperm (void ) {struct selinux_state *state = &selinux_state; -return state->policycap[POLICYDB_CAPABILITY_OPENPERM ]; +return READ_ONCE (state->policycap[POLICYDB_CAPABILITY_OPENPERM ]); }static inline bool selinux_policycap_extsockclass (void ) {struct selinux_state *state = &selinux_state; -return state->policycap[POLICYDB_CAPABILITY_EXTSOCKCLASS ]; +return READ_ONCE (state->policycap[POLICYDB_CAPABILITY_EXTSOCKCLASS ]); }static inline bool selinux_policycap_alwaysnetwork (void ) {struct selinux_state *state = &selinux_state; -return state->policycap[POLICYDB_CAPABILITY_ALWAYSNETWORK ]; +return READ_ONCE (state->policycap[POLICYDB_CAPABILITY_ALWAYSNETWORK ]); }static inline bool selinux_policycap_cgroupseclabel (void ) {struct selinux_state *state = &selinux_state; -return state->policycap[POLICYDB_CAPABILITY_CGROUPSECLABEL ]; +return READ_ONCE (state->policycap[POLICYDB_CAPABILITY_CGROUPSECLABEL ]); }static inline bool selinux_policycap_nnp_nosuid_transition (void ) {struct selinux_state *state = &selinux_state; -return state->policycap[POLICYDB_CAPABILITY_NNP_NOSUID_TRANSITION ]; +return READ_ONCE (state->policycap[POLICYDB_CAPABILITY_NNP_NOSUID_TRANSITION ]); }static inline bool selinux_policycap_genfs_seclabel_symlinks (void ) {struct selinux_state *state = &selinux_state; -return state->policycap[POLICYDB_CAPABILITY_GENFS_SECLABEL_SYMLINKS ]; +return READ_ONCE (state->policycap[POLICYDB_CAPABILITY_GENFS_SECLABEL_SYMLINKS ]); }int security_mls_enabled (struct selinux_state *state);
security/selinux/ss/services.c +2 -1
@@ -2103,7 +2103,8 @@ static void security_load_policycaps(struct selinux_state *state)struct ebitmap_node *node;for (i =0 ; i <ARRAY_SIZE (state->policycap); i++) - state->policycap[i] =ebitmap_get_bit (&p->policycaps, i); +WRITE_ONCE (state->policycap[i], +ebitmap_get_bit (&p->policycaps, i));for (i =0 ; i <ARRAY_SIZE (selinux_policycap_names); i++) pr_info("SELinux: policy capability %s=%d\n",
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 69/147 samples/bpf: Fix possible deadlock in xdpsock
In reply to Sasha LevinFrom: Magnus Karlsson <[email protected]>
[ Upstream commit 5a2a0dd88f0f267ac5953acd81050ae43a82201f ]
Fix a possible deadlock in the l2fwd application in xdpsock that can occur when there is no space in the Tx ring. There are two ways to get the kernel to consume entries in the Tx ring: calling sendto() to make it send packets and freeing entries from the completion ring, as the kernel will not send a packet if there is no space for it to add a completion entry in the completion ring. The Tx loop in l2fwd only used to call sendto(). This patches adds cleaning the completion ring in that loop.
--- samples/bpf/xdpsock_user.c | 1 + 1 file changed, 1 insertion(+)
samples/bpf/xdpsock_user.c +1 -0
@@ -1111,6 +1111,7 @@ static void l2fwd(struct xsk_socket_info *xsk, struct pollfd *fds)while (ret != rcvd) {if (ret <0 )exit_with_error (-ret); +complete_tx_l2fwd (xsk, fds);if (xsk_ring_prod__needs_wakeup (&xsk->tx))kick_tx (xsk); ret =xsk_ring_prod__reserve (&xsk->tx, rcvd, &idx_tx);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 70/147 drm/amd/display: Check clock table return
In reply to Sasha LevinFrom: Rodrigo Siqueira <[email protected]>
[ Upstream commit 4b4f21ff7f5d11bb77e169b306dcbc5b216f5db5 ]
During the load processes for Renoir, our display code needs to retrieve the SMU clock and voltage table, however, this operation can fail which means that we have to check this scenario. Currently, we are not handling this case properly and as a result, we have seen the following dmesg log during the boot:
RIP: 0010:rn_clk_mgr_construct+0x129/0x3d0 [amdgpu] ... Call Trace: dc_clk_mgr_create+0x16a/0x1b0 [amdgpu] dc_create+0x231/0x760 [amdgpu]
This commit fixes this issue by checking the return status retrieved from the clock table before try to populate any bandwidth.
--- drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +5 -2
@@ -761,6 +761,7 @@ void rn_clk_mgr_construct( {struct dc_debug_options *debug = &ctx->dc->debug;struct dpm_clocks clock_table = {0 }; +enum pp_smu_status status =0 ; clk_mgr->base.ctx = ctx; clk_mgr->base.funcs = &dcn21_funcs;
@@ -817,8 +818,10 @@ void rn_clk_mgr_construct( clk_mgr->base.bw_params = &rn_bw_params;if (pp_smu && pp_smu->rn_funcs.get_dpm_clock_table) { - pp_smu->rn_funcs.get_dpm_clock_table (&pp_smu->rn_funcs.pp_smu, &clock_table); -if (ctx->dc_bios && ctx->dc_bios->integrated_info) { + status = pp_smu->rn_funcs.get_dpm_clock_table (&pp_smu->rn_funcs.pp_smu, &clock_table); + +if (status ==PP_SMU_RESULT_OK && + ctx->dc_bios && ctx->dc_bios->integrated_info) {rn_clk_mgr_helper_populate_bw_params (clk_mgr->base.bw_params, &clock_table, ctx->dc_bios->integrated_info); } }
Signature · 6 lines
2.25.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 71/147 riscv: Define AT_VECTOR_SIZE_ARCH for ARCH_DLINFO
In reply to Sasha LevinFrom: Zong Li <[email protected]>
[ Upstream commit b5fca7c55f9fbab5ad732c3bce00f31af6ba5cfa ]
AT_VECTOR_SIZE_ARCH should be defined with the maximum number of NEW_AUX_ENT entries that ARCH_DLINFO can contain, but it wasn't defined for RISC-V at all even though ARCH_DLINFO will contain one NEW_AUX_ENT for the VDSO address.
--- arch/riscv/include/uapi/asm/auxvec.h | 3 +++ 1 file changed, 3 insertions(+)
arch/riscv/include/uapi/asm/auxvec.h +3 -0
@@ -10,4 +10,7 @@/* vDSO location */ #define AT_SYSINFO_EHDR 33 +/* entries in ARCH_DLINFO */ +#define AT_VECTOR_SIZE_ARCH 1 +#endif /* _UAPI_ASM_RISCV_AUXVEC_H */
Signature · 7 lines
2.25.1 _______________________________________________ linux-riscv mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-riscv
Sasha Levin
maintainerPATCH 72/147 cpufreq: sti-cpufreq: add stih418 support
In reply to Sasha LevinFrom: Alain Volmat <[email protected]>
[ Upstream commit 01a163c52039e9426c7d3d3ab16ca261ad622597 ]
The STiH418 can be controlled the same way as STiH407 & STiH410 regarding cpufreq.
--- drivers/cpufreq/sti-cpufreq.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/cpufreq/sti-cpufreq.c +4 -2
@@ -141,7 +141,8 @@ static const struct reg_field sti_stih407_dvfs_regfields[DVFS_MAX_REGFIELDS] = {static const struct reg_field *sti_cpufreq_match (void ) {if (of_machine_is_compatible ("st,stih407" ) || -of_machine_is_compatible ("st,stih410" )) +of_machine_is_compatible ("st,stih410" ) || +of_machine_is_compatible ("st,stih418" ))return sti_stih407_dvfs_regfields;return NULL ;
@@ -258,7 +259,8 @@ static int sti_cpufreq_init(void)int ret;if ((!of_machine_is_compatible ("st,stih407" )) && - (!of_machine_is_compatible ("st,stih410" ))) + (!of_machine_is_compatible ("st,stih410" )) && + (!of_machine_is_compatible ("st,stih418" )))return -ENODEV ; ddata.cpu =get_cpu_device (0 );
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 73/147 USB: adutux: fix debugging
In reply to Sasha LevinFrom: Oliver Neukum <[email protected]>
[ Upstream commit c56150c1bc8da5524831b1dac2eec3c67b89f587 ]
Handling for removal of the controller was missing at one place. Add it.
--- drivers/usb/misc/adutux.c | 1 + 1 file changed, 1 insertion(+)
drivers/usb/misc/adutux.c +1 -0
@@ -209,6 +209,7 @@ static void adu_interrupt_out_callback(struct urb *urb)if (status !=0 ) {if ((status != -ENOENT ) && + (status != -ESHUTDOWN ) && (status != -ECONNRESET )) { dev_dbg(&dev->udev->dev, "%s :nonzero status received: %d\n", __func__,
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 74/147 uio: free uio id after uio file node is freed
In reply to Sasha LevinFrom: Lang Dai <[email protected]>
[ Upstream commit 8fd0e2a6df262539eaa28b0a2364cca10d1dc662 ]
uio_register_device() do two things.
1) get an uio id from a global pool, e.g. the id is <A>
2) create file nodes like /sys/class/uio/uio<A>
uio_unregister_device() do two things.
1) free the uio id <A> and return it to the global pool
2) free the file node /sys/class/uio/uio<A>
There is a situation is that one worker is calling uio_unregister_device(), and another worker is calling uio_register_device().
If the two workers are X and Y, they go as below sequence,
1) X free the uio id <AAA>
2) Y get an uio id <AAA>
3) Y create file node /sys/class/uio/uio<AAA>
4) X free the file note /sys/class/uio/uio<AAA>
Then it will failed at the 3rd step and cause the phenomenon we saw as it is creating a duplicated file node.
Failure reports as follows:
sysfs: cannot create duplicate filename '/class/uio/uio10'
Call Trace: sysfs_do_create_link_sd.isra.2+0x9e/0xb0 sysfs_create_link+0x25/0x40 device_add+0x2c4/0x640 __uio_register_device+0x1c5/0x576 [uio] adf_uio_init_bundle_dev+0x231/0x280 [intel_qat] adf_uio_register+0x1c0/0x340 [intel_qat] adf_dev_start+0x202/0x370 [intel_qat] adf_dev_start_async+0x40/0xa0 [intel_qat] process_one_work+0x14d/0x410 worker_thread+0x4b/0x460 kthread+0x105/0x140 ? process_one_work+0x410/0x410 ? kthread_bind+0x40/0x40 ret_from_fork+0x1f/0x40 Code: 85 c0 48 89 c3 74 12 b9 00 10 00 00 48 89 c2 31 f6 4c 89 ef e8 ec c4 ff ff 4c 89 e2 48 89 de 48 c7 c7 e8 b4 ee b4 e8 6a d4 d7 ff <0f> 0b 48 89 df e8 20 fa f3 ff 5b 41 5c 41 5d 5d c3 66 0f 1f 84 ---[ end trace a7531c1ed5269e84 ]--- c6xxvf b002:00:00.0: Failed to register UIO devices c6xxvf b002:00:00.0: Failed to register UIO devices
--- drivers/uio/uio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/uio/uio.c +2 -2
@@ -1048,8 +1048,6 @@ void uio_unregister_device(struct uio_info *info) idev = info->uio_dev; -uio_free_minor (idev); -mutex_lock (&idev->info_lock);uio_dev_del_attributes (idev);
@@ -1064,6 +1062,8 @@ void uio_unregister_device(struct uio_info *info)device_unregister (&idev->dev); +uio_free_minor (idev); +return ; }EXPORT_SYMBOL_GPL (uio_unregister_device);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 75/147 coresight: Make sysfs functional on topologies with per core sink
In reply to Sasha LevinFrom: Linu Cherian <[email protected]>
[ Upstream commit 6d578258b955fc8888e1bbd9a8fefe7b10065a84 ]
Coresight driver assumes sink is common across all the ETMs, and tries to build a path between ETM and the first enabled sink found using bus based search. This breaks sysFS usage on implementations that has multiple per core sinks in enabled state.
To fix this, coresight_get_enabled_sink API is updated to do a connection based search starting from the given source, instead of bus based search.
With sink selection using sysfs depecrated for perf interface, provision for reset is removed as well in this API.
--- drivers/hwtracing/coresight/coresight-priv.h | 3 +- drivers/hwtracing/coresight/coresight.c | 62 +++++++++----------- 2 files changed, 29 insertions(+), 36 deletions(-)
drivers/hwtracing/coresight/coresight-priv.h +2 -1
@@ -148,7 +148,8 @@ static inline void coresight_write_reg_pair(void __iomem *addr, u64 val,void coresight_disable_path (struct list_head *path);int coresight_enable_path (struct list_head *path,u32 mode,void *sink_data);struct coresight_device *coresight_get_sink (struct list_head *path); -struct coresight_device *coresight_get_enabled_sink (bool reset); +struct coresight_device * +coresight_get_enabled_sink (struct coresight_device *source);struct coresight_device *coresight_get_sink_by_id (u32 id);struct coresight_device *coresight_find_default_sink (struct coresight_device *csdev);
drivers/hwtracing/coresight/coresight.c +27 -35
88 lines not on this page · raw message
Signature · 7 lines
2.25.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Re: [PATCH AUTOSEL 5.9 075/147] coresight: Make sysfs functional on topologies with per core sink
Hi,
Upstream commit,
commit bb1860efc817c18fce4112f25f51043e44346d1b
Author: Linu Cherian <[email protected]> Date: Wed Sep 16 13:17:34 2020 -0600
coresight: etm: perf: Sink selection using sysfs is deprecated
need to go along with this, else there will be build breakage. This applies for 5.4, 5.8 and 5.9
Mathieu, could you please ACK ?
Please let me know if i need to send the patch to
[email protected] separately.
Thanks.
Quoted 143 lines · Sasha Levin
From: Linu Cherian <[email protected]>
[ Upstream commit
6d578258b955fc8888e1bbd9a8fefe7b10065a84]Coresight driver assumes sink is common across all the ETMs, and tries to build a path between ETM and the first enabled sink found using bus based search. This breaks sysFS usage on implementations that has multiple per core sinks in enabled state.
To fix this,
coresight_get_enabled_sinkAPI is updated to do a connection based search starting from the given source, instead of bus based search.
With sink selection using sysfs depecrated for perf interface, provision for reset is removed as well in this API.Signed-off-by Linu Cherian <[email protected]>[Fixed indentation problem and removed obsolete comment]Signed-off-by Mathieu Poirier <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- drivers/hwtracing/coresight/coresight-priv.h | 3 +- drivers/hwtracing/coresight/coresight.c | 62 +++++++++----------- 2 files changed, 29 insertions(+), 36 deletions(-)drivers/hwtracing/coresight/coresight-priv.h +2
-1@@ -148,7 +148,8 @@ static inline void coresight_write_reg_pair(void __iomem *addr, u64 val,void coresight_disable_path (struct list_head *path);int coresight_enable_path (struct list_head *path,u32 mode,void *sink_data);struct coresight_device *coresight_get_sink (struct list_head *path); -struct coresight_device *coresight_get_enabled_sink (bool reset); +struct coresight_device * +coresight_get_enabled_sink (struct coresight_device *source);struct coresight_device *coresight_get_sink_by_id (u32 id);struct coresight_device *coresight_find_default_sink (struct coresight_device *csdev);drivers/hwtracing/coresight/coresight.c +27
-36@@ -540,50 +540,46 @@ struct coresight_device *coresight_get_sink(struct list_head *path)return csdev; } -static int coresight_enabled_sink (struct device *dev,const void *data) +static struct coresight_device * +coresight_find_enabled_sink (struct coresight_device *csdev) { -const bool *reset = data; -struct coresight_device *csdev =to_coresight_device (dev); +int i; +struct coresight_device *sink;if ((csdev->type ==CORESIGHT_DEV_TYPE_SINK || csdev->type ==CORESIGHT_DEV_TYPE_LINKSINK ) && - csdev->activated) { -/* -* Now that we have a handle on the sink for this session, -* disable the sysFS "enable_sink" flag so that possible -* concurrent perf session that wish to use another sink don't -* trip on it. Doing so has no ramification for the current -* session. -*/ -if (*reset) - csdev->activated = false; + csdev->activated) +return csdev; -return 1 ; +/* +* Recursively explore each port found on this element. +*/ +for (i =0 ; i < csdev->pdata->nr_outport; i++) { +struct coresight_device *child_dev; + + child_dev = csdev->pdata->conns[i].child_dev; +if (child_dev) + sink =coresight_find_enabled_sink (child_dev); +if (sink) +return sink; } -return 0 ; +return NULL ; }/** -* coresight_get_enabled_sink - returns the first enabled sink found on the bus -* @deactivate: Whether the 'enable_sink' flag should be reset -* -* When operated from perf the deactivate parameter should be set to 'true'. -* That way the "enabled_sink" flag of the sink that was selected can be reset, -* allowing for other concurrent perf sessions to choose a different sink. +* coresight_get_enabled_sink - returns the first enabled sink using +* connection based search starting from the source reference * -* When operated from sysFS users have full control and as such the deactivate -* parameter should be set to 'false', hence mandating users to explicitly -* clear the flag. +* @source: Coresight source device reference */ -struct coresight_device *coresight_get_enabled_sink (bool deactivate) +struct coresight_device * +coresight_get_enabled_sink (struct coresight_device *source) { -struct device *dev =NULL ; - - dev =bus_find_device (&coresight_bustype,NULL , &deactivate, - coresight_enabled_sink); +if (!source) +return NULL ; -return dev ?to_coresight_device (dev) :NULL ; +return coresight_find_enabled_sink (source); }static int coresight_sink_by_id (struct device *dev,const void *data)@@ -988,11 +984,7 @@ int coresight_enable(struct coresight_device *csdev) goto out; } -/* -* Search for a valid sink for this session but don't reset the -* "enable_sink" flag in sysFS. Users get to do that explicitly. -*/ - sink =coresight_get_enabled_sink (false); + sink =coresight_get_enabled_sink (csdev);if (!sink) { ret = -EINVAL ; goto out; --2.25.1
_______________________________________________
CoreSight mailing list
[email protected]
https://lists.linaro.org/mailman/listinfo/coresight
Re: [PATCH AUTOSEL 5.9 075/147] coresight: Make sysfs functional on topologies with per core sink
Quoted 11 lines · Linu Cherian
Hi,
Upstream commit,
commit
bb1860efc817c18fce4112f25f51043e44346d1bAuthor: Linu Cherian <[email protected]> Date: Wed Sep 16 13:17:34 2020 -0600
Quoted 7 lines
coresight: etm: perf: Sink selection using sysfs is deprecated
need to go along with this, else there will be build breakage. This applies for 5.4, 5.8 and 5.9
Mathieu, could you please ACK ?
Top posting makes it very difficult to follow what is going on.
Based on your above comment Sasha has probably dropped this patch. The best way to proceed is likely to send a patchset to stable with all the required patches.
Thanks,
Mathieu
Quoted 152 lines
Please let me know if i need to send the patch to
[email protected] separately.
Thanks.Quoted 143 lines · Sasha Levin
From: Linu Cherian <[email protected]>
[ Upstream commit
6d578258b955fc8888e1bbd9a8fefe7b10065a84]Coresight driver assumes sink is common across all the ETMs, and tries to build a path between ETM and the first enabled sink found using bus based search. This breaks sysFS usage on implementations that has multiple per core sinks in enabled state.
To fix this,
coresight_get_enabled_sinkAPI is updated to do a connection based search starting from the given source, instead of bus based search.
With sink selection using sysfs depecrated for perf interface, provision for reset is removed as well in this API.Signed-off-by Linu Cherian <[email protected]>[Fixed indentation problem and removed obsolete comment]Signed-off-by Mathieu Poirier <[email protected]>Signed-off-by Greg Kroah-Hartman <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- drivers/hwtracing/coresight/coresight-priv.h | 3 +- drivers/hwtracing/coresight/coresight.c | 62 +++++++++----------- 2 files changed, 29 insertions(+), 36 deletions(-)drivers/hwtracing/coresight/coresight-priv.h +2
-1@@ -148,7 +148,8 @@ static inline void coresight_write_reg_pair(void __iomem *addr, u64 val,void coresight_disable_path (struct list_head *path);int coresight_enable_path (struct list_head *path,u32 mode,void *sink_data);struct coresight_device *coresight_get_sink (struct list_head *path); -struct coresight_device *coresight_get_enabled_sink (bool reset); +struct coresight_device * +coresight_get_enabled_sink (struct coresight_device *source);struct coresight_device *coresight_get_sink_by_id (u32 id);struct coresight_device *coresight_find_default_sink (struct coresight_device *csdev);drivers/hwtracing/coresight/coresight.c +27
-36@@ -540,50 +540,46 @@ struct coresight_device *coresight_get_sink(struct list_head *path)return csdev; } -static int coresight_enabled_sink (struct device *dev,const void *data) +static struct coresight_device * +coresight_find_enabled_sink (struct coresight_device *csdev) { -const bool *reset = data; -struct coresight_device *csdev =to_coresight_device (dev); +int i; +struct coresight_device *sink;if ((csdev->type ==CORESIGHT_DEV_TYPE_SINK || csdev->type ==CORESIGHT_DEV_TYPE_LINKSINK ) && - csdev->activated) { -/* -* Now that we have a handle on the sink for this session, -* disable the sysFS "enable_sink" flag so that possible -* concurrent perf session that wish to use another sink don't -* trip on it. Doing so has no ramification for the current -* session. -*/ -if (*reset) - csdev->activated = false; + csdev->activated) +return csdev; -return 1 ; +/* +* Recursively explore each port found on this element. +*/ +for (i =0 ; i < csdev->pdata->nr_outport; i++) { +struct coresight_device *child_dev; + + child_dev = csdev->pdata->conns[i].child_dev; +if (child_dev) + sink =coresight_find_enabled_sink (child_dev); +if (sink) +return sink; } -return 0 ; +return NULL ; }/** -* coresight_get_enabled_sink - returns the first enabled sink found on the bus -* @deactivate: Whether the 'enable_sink' flag should be reset -* -* When operated from perf the deactivate parameter should be set to 'true'. -* That way the "enabled_sink" flag of the sink that was selected can be reset, -* allowing for other concurrent perf sessions to choose a different sink. +* coresight_get_enabled_sink - returns the first enabled sink using +* connection based search starting from the source reference * -* When operated from sysFS users have full control and as such the deactivate -* parameter should be set to 'false', hence mandating users to explicitly -* clear the flag. +* @source: Coresight source device reference */ -struct coresight_device *coresight_get_enabled_sink (bool deactivate) +struct coresight_device * +coresight_get_enabled_sink (struct coresight_device *source) { -struct device *dev =NULL ; - - dev =bus_find_device (&coresight_bustype,NULL , &deactivate, - coresight_enabled_sink); +if (!source) +return NULL ; -return dev ?to_coresight_device (dev) :NULL ; +return coresight_find_enabled_sink (source); }static int coresight_sink_by_id (struct device *dev,const void *data)@@ -988,11 +984,7 @@ int coresight_enable(struct coresight_device *csdev) goto out; } -/* -* Search for a valid sink for this session but don't reset the -* "enable_sink" flag in sysFS. Users get to do that explicitly. -*/ - sink =coresight_get_enabled_sink (false); + sink =coresight_get_enabled_sink (csdev);if (!sink) { ret = -EINVAL ; goto out; --2.25.1
_______________________________________________
CoreSight mailing list
[email protected]
https://lists.linaro.org/mailman/listinfo/coresight
Sasha Levin
maintainerPATCH 76/147 drm/amdgpu: No sysfs, not an error condition
In reply to Sasha LevinFrom: Luben Tuikov <[email protected]>
[ Upstream commit 5aea5327ea2ddf544cbeff096f45fc2319b0714e ]
Not being able to create amdgpu sysfs attributes is not a fatal error warranting not to continue to try to bring up the display. Thus, if we get an error trying to create amdgpu sysfs attrs, report it and continue on to try to bring up
a display.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +1 -3
@@ -3316,10 +3316,8 @@ int amdgpu_device_init(struct amdgpu_device *adev,flush_delayed_work (&adev->delayed_init_work); r =sysfs_create_files (&adev->dev->kobj, amdgpu_dev_attributes); -if (r) { +if (r)dev_err (adev->dev,"Could not create amdgpu device attr\n" ); -return r; - }if (IS_ENABLED (CONFIG_PERF_EVENTS )) r =amdgpu_pmu_init (adev);
Signature · 6 lines
2.25.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 77/147 mac80211: add missing queue/hash initialization to 802.3 xmit
In reply to Sasha LevinFrom: Felix Fietkau <[email protected]>
[ Upstream commit 5f8d69eaab1915df97f4f2aca89ea16abdd092d5 ]
Fixes AQL for encap-offloaded tx
--- net/mac80211/tx.c | 6 ++++++ 1 file changed, 6 insertions(+)
net/mac80211/tx.c +6 -0
@@ -4209,6 +4209,12 @@ static void ieee80211_8023_xmit(struct ieee80211_sub_if_data *sdata,if (is_zero_ether_addr (ra)) goto out_free; +if (local->ops->wake_tx_queue) { +u16 queue =__ieee80211_select_queue (sdata, sta, skb); +skb_set_queue_mapping (skb, queue); +skb_get_hash (skb); + } + multicast =is_multicast_ether_addr (ra);if (sta)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 78/147 usb: xhci: omit duplicate actions when suspending a runtime suspended host.
In reply to Sasha LevinFrom: Peter Chen <[email protected]>
[ Upstream commit 18a367e8947d72dd91b6fc401e88a2952c6363f7 ]
If the xhci-plat.c is the platform driver, after the runtime pm is enabled, the xhci_suspend is called if nothing is connected on the port. When the system goes to suspend, it will call xhci_suspend again if USB wakeup is enabled.
Since the runtime suspend wakeup setting is not always the same as system suspend wakeup setting, eg, at runtime suspend we always need wakeup if the controller is in low power mode; but at system suspend, we may not need wakeup. So, we move the judgement after changing wakeup setting.
[commit message rewording -Mathias]
--- drivers/usb/host/xhci.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/usb/host/xhci.c +5 -2
@@ -982,12 +982,15 @@ int xhci_suspend(struct xhci_hcd *xhci, bool do_wakeup) xhci->shared_hcd->state != HC_STATE_SUSPENDED)return -EINVAL ; -xhci_dbc_suspend (xhci); -/* Clear root port wake on bits if wakeup not allowed. */ if (!do_wakeup)xhci_disable_port_wake_on_bits (xhci); +if (!HCD_HW_ACCESSIBLE (hcd)) +return 0 ; + +xhci_dbc_suspend (xhci); +/* Don't poll the roothubs on bus suspend. */ xhci_dbg (xhci,"%s: stopping port polling.\n" , __func__);clear_bit (HCD_FLAG_POLL_RH , &hcd->flags);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 79/147 SUNRPC: Mitigate cond_resched() in xprt_transmit()
In reply to Sasha LevinFrom: Chuck Lever <[email protected]>
[ Upstream commit 6f9f17287e78e5049931af2037b15b26d134a32a ]
The original purpose of this expensive call is to prevent a long queue of requests from blocking other work.
The cond_resched() call is unnecessary after just a single send operation.
For longer queues, instead of invoking the kernel scheduler, simply release the transport send lock and return to the RPC scheduler.
--- net/sunrpc/xprt.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
net/sunrpc/xprt.c +4 -2
@@ -1520,10 +1520,13 @@ xprt_transmit(struct rpc_task *task) {struct rpc_rqst *next, *req = task->tk_rqstp;struct rpc_xprt *xprt = req->rq_xprt; -int status; +int counter, status;spin_lock (&xprt->queue_lock); + counter =0 ;while (!list_empty (&xprt->xmit_queue)) { +if (++counter ==20 ) +break ; next =list_first_entry (&xprt->xmit_queue, struct rpc_rqst, rq_xmit);xprt_pin_rqst (next);
@@ -1531,7 +1534,6 @@ xprt_transmit(struct rpc_task *task) status =xprt_request_transmit (next, task);if (status == -EBADMSG && next != req) status =0 ; -cond_resched ();spin_lock (&xprt->queue_lock);xprt_unpin_rqst (next);if (status ==0 ) {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 80/147 cpuidle: tegra: Correctly handle result of arm_cpuidle_simple_enter()
In reply to Sasha LevinFrom: Dmitry Osipenko <[email protected]>
[ Upstream commit 1170433e6611402b869c583fa1fbfd85106ff066 ]
The enter() callback of CPUIDLE drivers returns index of the entered idle state on success or a negative value on failure. The negative value could any negative value, i.e. it doesn't necessarily needs to be a error code. That's because CPUIDLE core only cares about the fact of failure and not about the reason of the enter() failure.
Like every other enter() callback, the arm_cpuidle_simple_enter() returns the entered idle-index on success. Unlike some of other drivers, it never fails. It happened that TEGRA_C1=index=err=0 in the code of cpuidle-tegra driver, and thus, there is no problem for the cpuidle-tegra driver created by the typo in the code which assumes that the arm_cpuidle_simple_enter() returns a error code.
The arm_cpuidle_simple_enter() also may return a -ENODEV error if CPU_IDLE is disabled in a kernel's config, but all CPUIDLE drivers are disabled if CPU_IDLE is disabled, including the cpuidle-tegra driver. So we can't ever see the error code from arm_cpuidle_simple_enter() today.
Of course the code may get some changes in the future and then the typo may transform into a real bug, so let's correct the typo! The tegra_cpuidle_state_enter() is now changed to make it return the entered idle-index on success and negative error code on fail, which puts it on par with the arm_cpuidle_simple_enter(), making code consistent in regards to the error handling.
This patch fixes a minor typo in the code, it doesn't fix any bugs.
--- drivers/cpuidle/cpuidle-tegra.c | 34 +++++++++++++++++++-------------- 1 file changed, 20 insertions(+), 14 deletions(-)
drivers/cpuidle/cpuidle-tegra.c +20 -14
@@ -172,7 +172,7 @@ static int tegra_cpuidle_coupled_barrier(struct cpuidle_device *dev)static int tegra_cpuidle_state_enter (struct cpuidle_device *dev,int index,unsigned int cpu) { -int ret; +int err;/* * CC6 state is the "CPU cluster power-off" state. In order to
@@ -183,9 +183,9 @@ static int tegra_cpuidle_state_enter(struct cpuidle_device *dev,* CPU cores, GIC and L2 cache). */ if (index ==TEGRA_CC6 ) { - ret =tegra_cpuidle_coupled_barrier (dev); -if (ret) -return ret; + err =tegra_cpuidle_coupled_barrier (dev); +if (err) +return err; }local_fiq_disable ();
@@ -194,15 +194,15 @@ static int tegra_cpuidle_state_enter(struct cpuidle_device *dev,switch (index) {case TEGRA_C7 : - ret =tegra_cpuidle_c7_enter (); + err =tegra_cpuidle_c7_enter ();break ;case TEGRA_CC6 : - ret =tegra_cpuidle_cc6_enter (cpu); + err =tegra_cpuidle_cc6_enter (cpu);break ;default : - ret = -EINVAL ; + err = -EINVAL ;break ; }
@@ -210,7 +210,7 @@ static int tegra_cpuidle_state_enter(struct cpuidle_device *dev,tegra_pm_clear_cpu_in_lp2 ();local_fiq_enable (); -return ret; +return err ?: index; }static int tegra_cpuidle_adjust_state_index (int index,unsigned int cpu)
@@ -236,21 +236,27 @@ static int tegra_cpuidle_enter(struct cpuidle_device *dev, int index) {unsigned int cpu =cpu_logical_map (dev->cpu); -int err; +int ret; index =tegra_cpuidle_adjust_state_index (index, cpu);if (dev->states_usage[index].disable)return -1 ;if (index ==TEGRA_C1 ) - err =arm_cpuidle_simple_enter (dev, drv, index); + ret =arm_cpuidle_simple_enter (dev, drv, index);else - err =tegra_cpuidle_state_enter (dev, index, cpu); + ret =tegra_cpuidle_state_enter (dev, index, cpu); -if (err && (err != -EINTR || index !=TEGRA_CC6 )) -pr_err_once ("failed to enter state %d err: %d\n" , index, err); +if (ret <0 ) { +if (ret != -EINTR || index !=TEGRA_CC6 ) +pr_err_once ("failed to enter state %d err: %d\n" , + index, ret); + index =-1 ; + }else { + index = ret; + } -return err ?-1 : index; +return index; } static int tegra114_enter_s2idle(struct cpuidle_device *dev,
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 81/147 arm64/mm: return cpu_all_mask when node is NUMA_NO_NODE
In reply to Sasha LevinFrom: Zhengyuan Liu <[email protected]>
[ Upstream commit a194c5f2d2b3a05428805146afcabe5140b5d378 ]
The @node passed to cpumask_of_node() can be NUMA_NO_NODE, in that case it will trigger the following WARN_ON(node >= nr_node_ids) due to mismatched data types of @node and @nr_node_ids. Actually we should return cpu_all_mask just like most other architectures do if passed NUMA_NO_NODE.
Also add a similar check to the inline cpumask_of_node() in numa.h.
--- arch/arm64/include/asm/numa.h | 3 +++ arch/arm64/mm/numa.c | 6 +++++- 2 files changed, 8 insertions(+), 1 deletion(-)
arch/arm64/include/asm/numa.h +3 -0
@@ -25,6 +25,9 @@ const struct cpumask *cpumask_of_node(int node);/* Returns a pointer to the cpumask of CPUs on Node 'node'. */ static inline const struct cpumask *cpumask_of_node (int node) { +if (node ==NUMA_NO_NODE ) +return cpu_all_mask; +return node_to_cpumask_map[node]; }#endif
arch/arm64/mm/numa.c +5 -1
@@ -46,7 +46,11 @@ EXPORT_SYMBOL(node_to_cpumask_map);*/ const struct cpumask *cpumask_of_node (int node) { -if (WARN_ON (node >= nr_node_ids)) + +if (node ==NUMA_NO_NODE ) +return cpu_all_mask; + +if (WARN_ON (node <0 || node >= nr_node_ids))return cpu_none_mask;if (WARN_ON (node_to_cpumask_map[node] ==NULL ))
Signature · 7 lines
2.25.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Sasha Levin
maintainerPATCH 82/147 can: flexcan: disable clocks during stop mode
In reply to Sasha LevinFrom: Joakim Zhang <[email protected]>
[ Upstream commit 02f71c6605e1f8259c07f16178330db766189a74 ]
Disable clocks while CAN core is in stop mode.
--- drivers/net/can/flexcan.c | 30 ++++++++++++++++++++---------- 1 file changed, 20 insertions(+), 10 deletions(-)
drivers/net/can/flexcan.c +20 -10
@@ -1700,8 +1700,6 @@ static int __maybe_unused flexcan_suspend(struct device *device) err =flexcan_chip_disable (priv);if (err)return err; - - err =pm_runtime_force_suspend (device); }netif_stop_queue (dev);netif_device_detach (dev);
@@ -1727,10 +1725,6 @@ static int __maybe_unused flexcan_resume(struct device *device)if (err)return err; }else { - err =pm_runtime_force_resume (device); -if (err) -return err; - err =flexcan_chip_enable (priv); } }
@@ -1761,8 +1755,16 @@ static int __maybe_unused flexcan_noirq_suspend(struct device *device)struct net_device *dev =dev_get_drvdata (device);struct flexcan_priv *priv =netdev_priv (dev); -if (netif_running (dev) &&device_may_wakeup (device)) -flexcan_enable_wakeup_irq (priv, true); +if (netif_running (dev)) { +int err; + +if (device_may_wakeup (device)) +flexcan_enable_wakeup_irq (priv, true); + + err =pm_runtime_force_suspend (device); +if (err) +return err; + }return 0 ; }
@@ -1772,8 +1774,16 @@ static int __maybe_unused flexcan_noirq_resume(struct device *device)struct net_device *dev =dev_get_drvdata (device);struct flexcan_priv *priv =netdev_priv (dev); -if (netif_running (dev) &&device_may_wakeup (device)) -flexcan_enable_wakeup_irq (priv, false); +if (netif_running (dev)) { +int err; + + err =pm_runtime_force_resume (device); +if (err) +return err; + +if (device_may_wakeup (device)) +flexcan_enable_wakeup_irq (priv, false); + }return 0 ; }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 83/147 habanalabs: remove security from ARB_MST_QUIET register
In reply to Sasha LevinFrom: farah kassabri <[email protected]>
[ Upstream commit acd330c141b4c49f468f00719ebc944656061eac ]
Allow user application to write to this register in order to be able to configure the quiet period of the QMAN between grants.
--- .../misc/habanalabs/gaudi/gaudi_security.c | 55 +++++++------------ 1 file changed, 19 insertions(+), 36 deletions(-)
drivers/misc/habanalabs/gaudi/gaudi_security.c +19 -36
@@ -831,8 +831,7 @@ static void gaudi_init_mme_protection_bits(struct hl_device *hdev)PROT_BITS_OFFS ; word_offset = ((mmMME0_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmMME0_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmMME0_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmMME0_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmMME0_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmMME0_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmMME0_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -1311,8 +1310,7 @@ static void gaudi_init_mme_protection_bits(struct hl_device *hdev)PROT_BITS_OFFS ; word_offset = ((mmMME2_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmMME2_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmMME2_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmMME2_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmMME2_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmMME2_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmMME2_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -1790,8 +1788,7 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA0_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA0_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA0_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmDMA0_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA0_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA0_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA0_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -2186,8 +2183,7 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA1_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA1_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA1_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmDMA1_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA1_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA1_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA1_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -2582,8 +2578,7 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA2_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA2_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA2_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmDMA2_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA2_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA2_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA2_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -2978,8 +2973,7 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA3_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA3_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA3_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmDMA3_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA3_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA3_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA3_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -3374,8 +3368,7 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA4_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA4_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA4_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmDMA4_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA4_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA4_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA4_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -3770,8 +3763,7 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA5_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA5_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA5_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmDMA5_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA5_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA5_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA5_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -4166,8 +4158,8 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA6_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA6_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA6_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + + mask =1 << ((mmDMA6_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA6_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA6_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA6_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -4562,8 +4554,7 @@ static void gaudi_init_dma_protection_bits(struct hl_device *hdev) word_offset = ((mmDMA7_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmDMA7_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmDMA7_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmDMA7_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmDMA7_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmDMA7_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmDMA7_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -5491,8 +5482,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev) word_offset = ((mmTPC0_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC0_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC0_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC0_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC0_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC0_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC0_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -5947,8 +5937,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev) word_offset = ((mmTPC1_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC1_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC1_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC1_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC1_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC1_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC1_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -6402,8 +6391,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev)PROT_BITS_OFFS ; word_offset = ((mmTPC2_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC2_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC2_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC2_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC2_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC2_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC2_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -6857,8 +6845,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev)PROT_BITS_OFFS ; word_offset = ((mmTPC3_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC3_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC3_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC3_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC3_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC3_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC3_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -7312,8 +7299,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev)PROT_BITS_OFFS ; word_offset = ((mmTPC4_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC4_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC4_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC4_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC4_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC4_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC4_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -7767,8 +7753,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev)PROT_BITS_OFFS ; word_offset = ((mmTPC5_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC5_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC5_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC5_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC5_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC5_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC5_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -8223,8 +8208,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev) word_offset = ((mmTPC6_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC6_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC6_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC6_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC6_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC6_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC6_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
@@ -8681,8 +8665,7 @@ static void gaudi_init_tpc_protection_bits(struct hl_device *hdev)PROT_BITS_OFFS ; word_offset = ((mmTPC7_QM_ARB_MST_CHOISE_PUSH_OFST_23 &PROT_BITS_OFFS ) >>7 ) <<2 ; - mask =1 << ((mmTPC7_QM_ARB_MST_QUIET_PER &0x7F ) >>2 ); - mask |=1 << ((mmTPC7_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); + mask =1 << ((mmTPC7_QM_ARB_SLV_CHOISE_WDT &0x7F ) >>2 ); mask |=1 << ((mmTPC7_QM_ARB_MSG_MAX_INFLIGHT &0x7F ) >>2 ); mask |=1 << ((mmTPC7_QM_ARB_MSG_AWUSER_31_11 &0x7F ) >>2 ); mask |=1 << ((mmTPC7_QM_ARB_MSG_AWUSER_SEC_PROP &0x7F ) >>2 );
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 84/147 xfs: don't free rt blocks when we're doing a REMAP bunmapi call
In reply to Sasha LevinFrom: "Darrick J. Wong" <[email protected]>
[ Upstream commit 8df0fa39bdd86ca81a8d706a6ed9d33cc65ca625 ]
When callers pass XFS_BMAPI_REMAP into xfs_bunmapi, they want the extent to be unmapped from the given file fork without the extent being freed. We do this for non-rt files, but we forgot to do this for realtime files. So far this isn't a big deal since nobody makes a bunmapi call to a rt file with the REMAP flag set, but don't leave a logic bomb.
--- fs/xfs/libxfs/xfs_bmap.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-)
fs/xfs/libxfs/xfs_bmap.c +12 -7
@@ -5046,20 +5046,25 @@ xfs_bmap_del_extent_real( flags =XFS_ILOG_CORE ;if (whichfork ==XFS_DATA_FORK &&XFS_IS_REALTIME_INODE (ip)) { -xfs_fsblock_t bno;xfs_filblks_t len;xfs_extlen_t mod; - bno =div_u64_rem (del->br_startblock, mp->m_sb.sb_rextsize, - &mod); -ASSERT (mod ==0 ); len =div_u64_rem (del->br_blockcount, mp->m_sb.sb_rextsize, &mod);ASSERT (mod ==0 ); - error =xfs_rtfree_extent (tp, bno, (xfs_extlen_t )len); -if (error) - goto done; +if (!(bflags &XFS_BMAPI_REMAP )) { +xfs_fsblock_t bno; + + bno =div_u64_rem (del->br_startblock, + mp->m_sb.sb_rextsize, &mod); +ASSERT (mod ==0 ); + + error =xfs_rtfree_extent (tp, bno, (xfs_extlen_t )len); +if (error) + goto done; + } + do_fx =0 ; nblks = len * mp->m_sb.sb_rextsize; qfield =XFS_TRANS_DQ_RTBCOUNT ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 85/147 xfs: avoid LR buffer overrun due to crafted h_len
In reply to Sasha LevinFrom: Gao Xiang <[email protected]>
[ Upstream commit f692d09e9c8fd0f5557c2e87f796a16dd95222b8 ]
Currently, crafted h_len has been blocked for the log header of the tail block in commit a70f9fe52daa ("xfs:
detect and handle invalid iclog size set by mkfs").
However, each log record could still have crafted h_len and cause log record buffer overrun. So let's check h_len vs buffer size for each log record as well.
--- fs/xfs/xfs_log_recover.c | 39 +++++++++++++++++++-------------------- 1 file changed, 19 insertions(+), 20 deletions(-)
fs/xfs/xfs_log_recover.c +19 -20
@@ -2904,7 +2904,8 @@ STATIC intxlog_valid_rec_header ( struct xlog *log, struct xlog_rec_header *rhead, - xfs_daddr_t blkno) + xfs_daddr_t blkno, + int bufsize) { int hlen;
@@ -2920,10 +2921,14 @@ xlog_valid_rec_header(return -EFSCORRUPTED ; } -/* LR body must have data or it wouldn't have been written */ +/* +* LR body must have data (or it wouldn't have been written) +* and h_len must not be greater than LR buffer size. +*/ hlen =be32_to_cpu (rhead->h_len); -if (XFS_IS_CORRUPT (log->l_mp, hlen <=0 || hlen >INT_MAX )) +if (XFS_IS_CORRUPT (log->l_mp, hlen <=0 || hlen > bufsize))return -EFSCORRUPTED ; +if (XFS_IS_CORRUPT (log->l_mp, blkno > log->l_logBBsize || blkno >INT_MAX ))return -EFSCORRUPTED ;
@@ -2984,9 +2989,6 @@ xlog_do_recovery_pass( goto bread_err1; rhead = (xlog_rec_header_t *)offset; - error =xlog_valid_rec_header (log, rhead, tail_blk); -if (error) - goto bread_err1;/* * xfsprogs has a bug where record length is based on lsunit but
@@ -3001,21 +3003,18 @@ xlog_do_recovery_pass(*/ h_size =be32_to_cpu (rhead->h_size); h_len =be32_to_cpu (rhead->h_len); -if (h_len > h_size) { -if (h_len <= log->l_mp->m_logbsize && -be32_to_cpu (rhead->h_num_logops) ==1 ) { -xfs_warn (log->l_mp, +if (h_len > h_size && h_len <= log->l_mp->m_logbsize && + rhead->h_num_logops ==cpu_to_be32 (1 )) { +xfs_warn (log->l_mp,"invalid iclog size (%d bytes), using lsunit (%d bytes)" , - h_size, log->l_mp->m_logbsize); - h_size = log->l_mp->m_logbsize; - }else { -XFS_ERROR_REPORT (__func__,XFS_ERRLEVEL_LOW , - log->l_mp); - error = -EFSCORRUPTED ; - goto bread_err1; - } + h_size, log->l_mp->m_logbsize); + h_size = log->l_mp->m_logbsize; } + error =xlog_valid_rec_header (log, rhead, tail_blk, h_size); +if (error) + goto bread_err1; +if ((be32_to_cpu (rhead->h_version) &XLOG_VERSION_2 ) && (h_size >XLOG_HEADER_CYCLE_SIZE )) { hblks = h_size /XLOG_HEADER_CYCLE_SIZE ;
@@ -3096,7 +3095,7 @@ xlog_do_recovery_pass( } rhead = (xlog_rec_header_t *)offset; error =xlog_valid_rec_header (log, rhead, - split_hblks ? blk_no :0 ); + split_hblks ? blk_no :0 , h_size);if (error) goto bread_err2;
@@ -3177,7 +3176,7 @@ xlog_do_recovery_pass( goto bread_err2; rhead = (xlog_rec_header_t *)offset; - error =xlog_valid_rec_header (log, rhead, blk_no); + error =xlog_valid_rec_header (log, rhead, blk_no, h_size);if (error) goto bread_err2;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 86/147 ACPI: Add out of bounds and numa_off protections to pxm_to_node()
In reply to Sasha LevinFrom: Jonathan Cameron <[email protected]>
[ Upstream commit 8a3decac087aa897df5af04358c2089e52e70ac4 ]
The function should check the validity of the pxm value before using it to index the pxm_to_node_map[] array.
Whilst hardening this code may be good in general, the main intent here is to enable following patches that use this function to replace acpi_map_pxm_to_node() for non SRAT usecases which should return NO_NUMA_NODE for PXM entries not matching with those in SRAT.
--- drivers/acpi/numa/srat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/acpi/numa/srat.c +1 -1
@@ -31,7 +31,7 @@ int acpi_numa __initdata;int pxm_to_node (int pxm) { -if (pxm <0 ) +if (pxm <0 || pxm >=MAX_PXM_DOMAINS || numa_off)return NUMA_NO_NODE ;return pxm_to_node_map[pxm]; }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 87/147 genirq: Add stub for set_handle_irq() when !GENERIC_IRQ_MULTI_HANDLER
In reply to Sasha LevinFrom: Zhen Lei <[email protected]>
[ Upstream commit ea0c80d1764449acf2f70fdb25aec33800cd0348 ]
In order to avoid compilation errors when a driver references set_handle_irq(), but that the architecture doesn't select GENERIC_IRQ_MULTI_HANDLER, add a stub function that will just WARN_ON_ONCE() if ever used.
--- include/linux/irq.h | 6 ++++++ 1 file changed, 6 insertions(+)
include/linux/irq.h +6 -0
@@ -1252,6 +1252,12 @@ int __init set_handle_irq(void (*handle_irq)(struct pt_regs *));* top-level IRQ handler. */ extern void (*handle_arch_irq)(struct pt_regs *)__ro_after_init ; +#else +#define set_handle_irq (handle_irq) \ + do { \ + (void)handle_irq; \ + WARN_ON(1); \ + } while (0)#endif #endif /* _LINUX_IRQ_H */
Signature · 1 line
2.25.1
Re: [PATCH AUTOSEL 5.9 087/147] genirq: Add stub for set_handle_irq() when !GENERIC_IRQ_MULTI_HANDLER
Quoted 34 lines · Sasha Levin
From: Zhen Lei <[email protected]>
[ Upstream commit
ea0c80d1764449acf2f70fdb25aec33800cd0348]In order to avoid compilation errors when a driver references
set_handle_irq(), but that the architecture doesn't selectGENERIC_IRQ_MULTI_HANDLER, add a stub function that will justWARN_ON_ONCE()if ever used.Signed-off-by Zhen Lei <[email protected]>[maz: commit message]Signed-off-by Marc Zyngier <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- include/linux/irq.h | 6 ++++++ 1 file changed, 6 insertions(+)include/linux/irq.h +6
-0@@ -1252,6 +1252,12 @@ int __init set_handle_irq(void(*handle_irq)(struct pt_regs *)); * top-level IRQ handler. */ extern void (*handle_arch_irq)(struct pt_regs *)__ro_after_init ; +#else +#define set_handle_irq (handle_irq) \ + do { \ + (void)handle_irq; \ + WARN_ON(1); \ + } while (0)#endif
Quoted 1 line
#endif /*
_LINUX_IRQ_H*/
What is the reason for this backport? The only user is a driver that isn't getting backported (d59f7d159891 and following patches).
Thanks,
M.
Signature · 1 line
Jazz is not dead. It just smells funny...
Sasha Levin
maintainerRe: [PATCH AUTOSEL 5.9 087/147] genirq: Add stub for set_handle_irq() when !GENERIC_IRQ_MULTI_HANDLER
Quoted 35 lines · Marc Zyngier
Quoted 34 lines · Sasha Levin
From: Zhen Lei <[email protected]>
[ Upstream commit
ea0c80d1764449acf2f70fdb25aec33800cd0348]In order to avoid compilation errors when a driver references
set_handle_irq(), but that the architecture doesn't selectGENERIC_IRQ_MULTI_HANDLER, add a stub function that will justWARN_ON_ONCE()if ever used.Signed-off-by Zhen Lei <[email protected]>[maz: commit message]Signed-off-by Marc Zyngier <[email protected]>Signed-off-by Sasha Levin <[email protected]>---
include/linux/irq.h| 6 ++++++1 file changed, 6 insertions(+)include/linux/irq.h +6
-0@@ -1252,6 +1252,12 @@ int __init set_handle_irq(void(*handle_irq)(struct pt_regs *)); * top-level IRQ handler. */ extern void (*handle_arch_irq)(struct pt_regs *)__ro_after_init ; +#else +#define set_handle_irq (handle_irq) \ + do { \ + (void)handle_irq; \ + WARN_ON(1); \ + } while (0)#endif
Quoted 4 lines
Quoted 1 line
#endif /*
_LINUX_IRQ_H*/What is the reason for this backport? The only user is a driver that isn't getting backported (
d59f7d159891and following patches).
I'll drop it, thanks!
Signature · 2 lines
Thanks, Sasha
Sasha Levin
maintainerPATCH 88/147 octeontx2-af: fix LD CUSTOM LTYPE aliasing
In reply to Sasha LevinFrom: Stanislaw Kardach <[email protected]>
[ Upstream commit 450f0b978870c384dd81d1176088536555f3170e ]
Since LD contains LTYPE definitions tweaked toward efficient NIX_AF_RX_FLOW_KEY_ALG(0..31)_FIELD(0..4) usage, the original location of NPC_LT_LD_CUSTOM0/1 was aliased with MPLS_IN_* definitions. Moving custom frame to value 6 and 7 removes the aliasing at the cost of custom frames being also considered when TCP/UDP RSS algo is configured.
However since the goal of CUSTOM frames is to classify them to a separate set of RQs, this cost is acceptable.
--- drivers/net/ethernet/marvell/octeontx2/af/npc.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/marvell/octeontx2/af/npc.h +2 -2
@@ -77,6 +77,8 @@ enum npc_kpu_ld_ltype {NPC_LT_LD_ICMP ,NPC_LT_LD_SCTP ,NPC_LT_LD_ICMP6 , +NPC_LT_LD_CUSTOM0 , +NPC_LT_LD_CUSTOM1 ,NPC_LT_LD_IGMP =8 ,NPC_LT_LD_ESP , NPC_LT_LD_AH,
@@ -85,8 +87,6 @@ enum npc_kpu_ld_ltype {NPC_LT_LD_NSH ,NPC_LT_LD_TU_MPLS_IN_NSH , NPC_LT_LD_TU_MPLS_IN_IP, -NPC_LT_LD_CUSTOM0 =0xE , - NPC_LT_LD_CUSTOM1 = 0xF, }; enum npc_kpu_le_ltype {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 89/147 dm: change max_io_len() to use blk_max_size_offset()
In reply to Sasha LevinFrom: Mike Snitzer <[email protected]>
[ Upstream commit 5091cdec56faeaefa79de4b6cb3c3c55e50d1ac3 ]
Using blk_max_size_offset() enables DM core's splitting to impose ti->max_io_len (via q->limits.chunk_sectors) and also fallback to respecting q->limits.max_sectors if chunk_sectors isn't set.
--- drivers/md/dm.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-)
drivers/md/dm.c +8 -12
@@ -1051,22 +1051,18 @@ static sector_t max_io_len_target_boundary(sector_t sector, struct dm_target *tistatic sector_t max_io_len (sector_t sector,struct dm_target *ti) {sector_t len =max_io_len_target_boundary (sector, ti); -sector_t offset, max_len; +sector_t max_len;/* * Does the target need to split even further? +* - q->limits.chunk_sectors reflects ti->max_io_len so +* blk_max_size_offset() provides required splitting. +* - blk_max_size_offset() also respects q->limits.max_sectors */ -if (ti->max_io_len) { - offset =dm_target_offset (ti, sector); -if (unlikely (ti->max_io_len & (ti->max_io_len -1 ))) - max_len =sector_div (offset, ti->max_io_len); -else - max_len = offset & (ti->max_io_len -1 ); - max_len = ti->max_io_len - max_len; - -if (len > max_len) - len = max_len; - } + max_len =blk_max_size_offset (dm_table_get_md (ti->table)->queue, +dm_target_offset (ti, sector)); +if (len > max_len) + len = max_len;return len; }
Signature · 7 lines
2.25.1 -- dm-devel mailing list [email protected] https://www.redhat.com/mailman/listinfo/dm-devel
Re: [PATCH AUTOSEL 5.9 089/147] dm: change max_io_len() to use blk_max_size_offset()
Quoted 10 lines · Sasha Levin
From: Mike Snitzer <[email protected]>
[ Upstream commit
5091cdec56faeaefa79de4b6cb3c3c55e50d1ac3]Using
blk_max_size_offset()enables DM core's splitting to imposeti->max_io_len(viaq->limits.chunk_sectors) and also fallback to respectingq->limits.max_sectorsifchunk_sectorsisn't set.Signed-off-by Mike Snitzer <[email protected]>Signed-off-by Sasha Levin <[email protected]>
Not sure why this commit elevated to stable@ picking it up, please explain.
But you cannot take this commit standalone. These commits are prereqs:
22ada802ede8 block: use lcm_not_zero() when stacking chunk_sectors 07d098e6bbad block: allow 'chunk_sectors' to be non-power-of-2882ec4e609c1 dm table: stack 'chunk_sectors' limit to account for target-specific splitting
This goes for all stable@ trees you AUTOSEL'd commit 5091cdec56f for.
Mike
Quoted 42 lines
--- drivers/md/dm.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-)drivers/md/dm.c +8
-12@@ -1051,22 +1051,18 @@ static sector_t max_io_len_target_boundary(sector_t sector, struct dm_target *tistatic sector_t max_io_len (sector_t sector,struct dm_target *ti) {sector_t len =max_io_len_target_boundary (sector, ti); -sector_t offset, max_len; +sector_t max_len;/* * Does the target need to split even further? +* - q->limits.chunk_sectors reflects ti->max_io_len so +* blk_max_size_offset() provides required splitting. +* - blk_max_size_offset() also respects q->limits.max_sectors */ -if (ti->max_io_len) { - offset =dm_target_offset (ti, sector); -if (unlikely (ti->max_io_len & (ti->max_io_len -1 ))) - max_len =sector_div (offset, ti->max_io_len); -else - max_len = offset & (ti->max_io_len -1 ); - max_len = ti->max_io_len - max_len; - -if (len > max_len) - len = max_len; - } + max_len =blk_max_size_offset (dm_table_get_md (ti->table)->queue, +dm_target_offset (ti, sector)); +if (len > max_len) + len = max_len;return len; }Signature · 1 line
2.25.1
Sasha Levin
maintainerRe: [PATCH AUTOSEL 5.9 089/147] dm: change max_io_len() to use blk_max_size_offset()
Quoted 11 lines · Mike Snitzer
Quoted 8 lines · Sasha Levin
From: Mike Snitzer <[email protected]>
[ Upstream commit
5091cdec56faeaefa79de4b6cb3c3c55e50d1ac3]Using
blk_max_size_offset()enables DM core's splitting to imposeti->max_io_len(viaq->limits.chunk_sectors) and also fallback to respectingq->limits.max_sectorsifchunk_sectorsisn't set.
Quoted 5 lines
Quoted 2 lines
Signed-off-by Mike Snitzer <[email protected]>Signed-off-by Sasha Levin <[email protected]>Not sure why this commit elevated to stable@ picking it up, please explain.
I misread the series this patch was in as being a fix rather than only the first patch, sorry :(
Quoted 53 lines
But you cannot take this commit standalone. These commits are prereqs:
22ada802ede8block: uselcm_not_zero()when stackingchunk_sectors07d098e6bbadblock: allow 'chunk_sectors' to be non-power-of-2882ec4e609c1dm table: stack 'chunk_sectors' limit to account for target-specific splittingThis goes for all stable@ trees you AUTOSEL'd commit
5091cdec56ffor.Mike
Quoted 42 lines
--- drivers/md/dm.c | 20 ++++++++------------ 1 file changed, 8 insertions(+), 12 deletions(-)drivers/md/dm.c +8
-13@@ -1051,22 +1051,18 @@ static sector_t max_io_len_target_boundary(sector_t sector, struct dm_target *tistatic sector_t max_io_len (sector_t sector,struct dm_target *ti) {sector_t len =max_io_len_target_boundary (sector, ti); -sector_t offset, max_len; +sector_t max_len;/* * Does the target need to split even further? +* - q->limits.chunk_sectors reflects ti->max_io_len so +* blk_max_size_offset() provides required splitting. +* - blk_max_size_offset() also respects q->limits.max_sectors */ -if (ti->max_io_len) { - offset =dm_target_offset (ti, sector); -if (unlikely (ti->max_io_len & (ti->max_io_len -1 ))) - max_len =sector_div (offset, ti->max_io_len); -else - max_len = offset & (ti->max_io_len -1 ); - max_len = ti->max_io_len - max_len; - -if (len > max_len) - len = max_len; - } + max_len =blk_max_size_offset (dm_table_get_md (ti->table)->queue, +dm_target_offset (ti, sector)); +if (len > max_len) + len = max_len;return len; } --2.25.1
Signature · 2 lines
Thanks, Sasha
Sasha Levin
maintainerPATCH 90/147 brcmfmac: Fix warning message after dongle setup failed
In reply to Sasha LevinFrom: Wright Feng <[email protected]>
[ Upstream commit 6aa5a83a7ed8036c1388a811eb8bdfa77b21f19c ]
Brcmfmac showed warning message in fweh.c when checking the size of event queue which is not initialized. Therefore, we only cancel the worker and reset event handler only when it is initialized.
[ 145.505899] brcmfmac 0000:02:00.0: brcmf_pcie_setup: Dongle setup [ 145.929970] ------------[ cut here ]------------ [ 145.929994] WARNING: CPU: 0 PID: 288 at drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c:312 brcmf_fweh_detach+0xbc/0xd0 [brcmfmac] ... [ 145.930029] Call Trace: [ 145.930036] brcmf_detach+0x77/0x100 [brcmfmac] [ 145.930043] brcmf_pcie_remove+0x79/0x130 [brcmfmac] [ 145.930046] pci_device_remove+0x39/0xc0 [ 145.930048] device_release_driver_internal+0x141/0x200 [ 145.930049] device_release_driver+0x12/0x20 [ 145.930054] brcmf_pcie_setup+0x101/0x3c0 [brcmfmac] [ 145.930060] brcmf_fw_request_done+0x11d/0x1f0 [brcmfmac] [ 145.930062] ? lock_timer_base+0x7d/0xa0 [ 145.930063] ? internal_add_timer+0x1f/0xa0 [ 145.930064] ? add_timer+0x11a/0x1d0 [ 145.930066] ? __kmalloc_track_caller+0x18c/0x230 [ 145.930068] ? kstrdup_const+0x23/0x30 [ 145.930069] ? add_dr+0x46/0x80 [ 145.930070] ? devres_add+0x3f/0x50 [ 145.930072] ? usermodehelper_read_unlock+0x15/0x20 [ 145.930073] ? _request_firmware+0x288/0xa20 [ 145.930075] request_firmware_work_func+0x36/0x60 [ 145.930077] process_one_work+0x144/0x360 [ 145.930078] worker_thread+0x4d/0x3c0 [ 145.930079] kthread+0x112/0x150 [ 145.930080] ? rescuer_thread+0x340/0x340 [ 145.930081] ? kthread_park+0x60/0x60 [ 145.930083] ret_from_fork+0x25/0x30
--- .../net/wireless/broadcom/brcm80211/brcmfmac/fweh.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/fweh.c +6 -4
@@ -304,10 +304,12 @@ void brcmf_fweh_detach(struct brcmf_pub *drvr) {struct brcmf_fweh_info *fweh = &drvr->fweh; -/* cancel the worker */ -cancel_work_sync (&fweh->event_work); -WARN_ON (!list_empty (&fweh->event_q)); -memset (fweh->evt_handler,0 ,sizeof (fweh->evt_handler)); +/* cancel the worker if initialized */ +if (fweh->event_work.func) { +cancel_work_sync (&fweh->event_work); +WARN_ON (!list_empty (&fweh->event_q)); +memset (fweh->evt_handler,0 ,sizeof (fweh->evt_handler)); + } }/**
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 91/147 ath11k: Use GFP_ATOMIC instead of GFP_KERNEL in ath11k_dp_htt_get_ppdu_desc
In reply to Sasha LevinFrom: Wen Gong <[email protected]>
[ Upstream commit 6a8be1baa9116a038cb4f6158cc10134387ca0d0 ]
With SLUB DEBUG CONFIG below crash is seen as kmem_cache_alloc is being called in non-atomic context.
To fix this issue, use GFP_ATOMIC instead of GFP_KERNEL kzalloc.
[ 357.217088] BUG: sleeping function called from invalid context at mm/slab.h:498 [ 357.217091] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 0, name: swapper/0 [ 357.217092] INFO: lockdep is turned off. [ 357.217095] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G W 5.9.0-rc5-wt-ath+ #196 [ 357.217096] Hardware name: Intel(R) Client Systems NUC8i7HVK/NUC8i7HVB, BIOS HNKBLi70.86A.0049.2018.0801.1601 08/01/2018 [ 357.217097] Call Trace: [ 357.217098] <IRQ> [ 357.217107] ? ath11k_dp_htt_get_ppdu_desc+0xa9/0x170 [ath11k] [ 357.217110] dump_stack+0x77/0xa0 [ 357.217113] ___might_sleep.cold+0xa6/0xb6 [ 357.217116] kmem_cache_alloc_trace+0x1f2/0x270 [ 357.217122] ath11k_dp_htt_get_ppdu_desc+0xa9/0x170 [ath11k] [ 357.217129] ath11k_htt_pull_ppdu_stats.isra.0+0x96/0x270 [ath11k] [ 357.217135] ath11k_dp_htt_htc_t2h_msg_handler+0xe7/0x1d0 [ath11k] [ 357.217137] ? trace_hardirqs_on+0x1c/0x100 [ 357.217143] ath11k_htc_rx_completion_handler+0x207/0x370 [ath11k] [ 357.217149] ath11k_ce_recv_process_cb+0x15e/0x1e0 [ath11k] [ 357.217151] ? handle_irq_event+0x70/0xa8 [ 357.217154] ath11k_pci_ce_tasklet+0x10/0x30 [ath11k_pci] [ 357.217157] tasklet_action_common.constprop.0+0xd4/0xf0 [ 357.217160] __do_softirq+0xc9/0x482 [ 357.217162] asm_call_on_stack+0x12/0x20 [ 357.217163] </IRQ> [ 357.217166] do_softirq_own_stack+0x49/0x60 [ 357.217167] irq_exit_rcu+0x9a/0xd0 [ 357.217169] common_interrupt+0xa1/0x190 [ 357.217171] asm_common_interrupt+0x1e/0x40 [ 357.217173] RIP: 0010:cpu_idle_poll.isra.0+0x2e/0x60 [ 357.217175] Code: 8b 35 26 27 74 69 e8 11 c8 3d ff e8 bc fa 42 ff e8 e7 9f 4a ff fb 65 48 8b 1c 25 80 90 01 00 48 8b 03 a8 08 74 0b eb 1c f3 90 <48> 8b 03 a8 08 75 13 8b 0 [ 357.217177] RSP: 0018:ffffffff97403ee0 EFLAGS: 00000202 [ 357.217178] RAX: 0000000000000001 RBX: ffffffff9742b8c0 RCX: 0000000000b890ca [ 357.217180] RDX: 0000000000b890ca RSI: 0000000000000001 RDI: ffffffff968d0c49 [ 357.217181] RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000001 [ 357.217182] R10: ffffffff9742b8c0 R11: 0000000000000046 R12: 0000000000000000 [ 357.217183] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000066fdf520 [ 357.217186] ? cpu_idle_poll.isra.0+0x19/0x60 [ 357.217189] do_idle+0x5f/0xe0 [ 357.217191] cpu_startup_entry+0x14/0x20 [ 357.217193] start_kernel+0x443/0x464 [ 357.217196] secondary_startup_64+0xa4/0xb0
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
--- drivers/net/wireless/ath/ath11k/dp_rx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/ath/ath11k/dp_rx.c +1 -1
@@ -1421,7 +1421,7 @@ struct htt_ppdu_stats_info *ath11k_dp_htt_get_ppdu_desc(struct ath11k *ar, }spin_unlock_bh (&ar->data_lock); - ppdu_info =kzalloc (sizeof (*ppdu_info),GFP_KERNEL ); + ppdu_info =kzalloc (sizeof (*ppdu_info),GFP_ATOMIC );if (!ppdu_info)return NULL ;
Signature · 7 lines
2.25.1 -- ath11k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath11k
Sasha Levin
maintainerPATCH 92/147 ath11k: fix warning caused by lockdep_assert_held
In reply to Sasha LevinFrom: Carl Huang <[email protected]>
[ Upstream commit 2f588660e34a982377109872757f1b99d7748d21 ]
Fix warning caused by lockdep_assert_held when CONFIG_LOCKDEP is enabled.
[ 271.940647] WARNING: CPU: 6 PID: 0 at drivers/net/wireless/ath/ath11k/hal.c:818 ath11k_hal_srng_access_begin+0x31/0x40 [ath11k] [ 271.940655] Modules linked in: qrtr_mhi qrtr ns ath11k_pci mhi ath11k qmi_helpers nvme nvme_core [ 271.940675] CPU: 6 PID: 0 Comm: swapper/6 Kdump: loaded Tainted: G W 5.9.0-rc5-kalle-bringup-wt-ath+ #4 [ 271.940682] Hardware name: Dell Inc. Inspiron 7590/08717F, BIOS 1.3.0 07/22/2019 [ 271.940698] RIP: 0010:ath11k_hal_srng_access_begin+0x31/0x40 [ath11k] [ 271.940708] Code: 48 89 f3 85 c0 75 11 48 8b 83 a8 00 00 00 8b 00 89 83 b0 00 00 00 5b c3 48 8d 7e 58 be ff ff ff ff e8 53 24 ec fa 85 c0 75 dd <0f> 0b eb d9 90 66 2e 0f 1f 84 00 00 00 00 00 55 53 48 89 f3 8b 35 [ 271.940718] RSP: 0018:ffffbdf0c0230df8 EFLAGS: 00010246 [ 271.940727] RAX: 0000000000000000 RBX: ffffa12b34e67680 RCX: ffffa12b57a0d800 [ 271.940735] RDX: 0000000000000000 RSI: 00000000ffffffff RDI: ffffa12b34e676d8 [ 271.940742] RBP: ffffa12b34e60000 R08: 0000000000000001 R09: 0000000000000001 [ 271.940753] R10: 0000000000000001 R11: 0000000000000046 R12: 0000000000000000 [ 271.940763] R13: ffffa12b34e60000 R14: ffffa12b34e60000 R15: 0000000000000000 [ 271.940774] FS: 0000000000000000(0000) GS:ffffa12b5a400000(0000) knlGS:0000000000000000 [ 271.940788] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 271.940798] CR2: 00007f8bef282008 CR3: 00000001f4224004 CR4: 00000000003706e0 [ 271.940805] Call Trace: [ 271.940813] <IRQ> [ 271.940835] ath11k_dp_tx_completion_handler+0x9e/0x950 [ath11k] [ 271.940847] ? lock_acquire+0xba/0x3b0 [ 271.940876] ath11k_dp_service_srng+0x5a/0x2e0 [ath11k] [ 271.940893] ath11k_pci_ext_grp_napi_poll+0x1e/0x80 [ath11k_pci] [ 271.940908] net_rx_action+0x283/0x4f0 [ 271.940931] __do_softirq+0xcb/0x499 [ 271.940950] asm_call_on_stack+0x12/0x20 [ 271.940963] </IRQ> [ 271.940979] do_softirq_own_stack+0x4d/0x60 [ 271.940991] irq_exit_rcu+0xb0/0xc0 [ 271.941001] common_interrupt+0xce/0x190 [ 271.941014] asm_common_interrupt+0x1e/0x40 [ 271.941026] RIP: 0010:cpuidle_enter_state+0x115/0x500
Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
--- drivers/net/wireless/ath/ath11k/dp_tx.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/net/wireless/ath/ath11k/dp_tx.c +4 -0
@@ -514,6 +514,8 @@ void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)u32 msdu_id;u8 mac_id; +spin_lock_bh (&status_ring->lock); +ath11k_hal_srng_access_begin (ab, status_ring); while ((ATH11K_TX_COMPL_NEXT(tx_ring->tx_status_head) !=
@@ -533,6 +535,8 @@ void ath11k_dp_tx_completion_handler(struct ath11k_base *ab, int ring_id)ath11k_hal_srng_access_end (ab, status_ring); +spin_unlock_bh (&status_ring->lock); +while (ATH11K_TX_COMPL_NEXT (tx_ring->tx_status_tail) != tx_ring->tx_status_head) {struct hal_wbm_release_ring *tx_status;u32 desc_id;
Signature · 7 lines
2.25.1 -- ath11k mailing list [email protected] http://lists.infradead.org/mailman/listinfo/ath11k
Sasha Levin
maintainerPATCH 93/147 ath11k: change to disable softirqs for ath11k_regd_update to solve deadlock
In reply to Sasha LevinFrom: Wen Gong <[email protected]>
[ Upstream commit df648808c6b9989555e247530d8ca0ad0094b361 ]
After base_lock which occupy by ath11k_regd_update, the softirq run for WMI_REG_CHAN_LIST_CC_EVENTID maybe arrived and it also need to accuire the spin lock, then deadlock happend, change to disable softirqis to solve it.
[ 235.576990] ================================
[ 235.576991] WARNING: inconsistent lock state
[ 235.576993] 5.9.0-rc5-wt-ath+ #196 Not tainted
[ 235.576994] --------------------------------
[ 235.576995] inconsistent {IN-SOFTIRQ-W} -> {SOFTIRQ-ON-W} usage.
[ 235.576997] kworker/u16:1/98 [HC0[0]:SC0[0]:HE1:SE1] takes:
[ 235.576998] ffff9655f75cad98 (&ab->base_lock){+.?.}-{2:2}, at: ath11k_regd_update+0x28/0x1d0 [ath11k]
[ 235.577009] {IN-SOFTIRQ-W} state was registered at:
[ 235.577013] __lock_acquire+0x219/0x6e0
[ 235.577015] lock_acquire+0xb6/0x270
[ 235.577018] _raw_spin_lock+0x2c/0x70
[ 235.577023] ath11k_reg_chan_list_event.isra.0+0x10d/0x1e0 [ath11k]
[ 235.577028] ath11k_wmi_tlv_op_rx+0x3c3/0x560 [ath11k]
[ 235.577033] ath11k_htc_rx_completion_handler+0x207/0x370 [ath11k]
[ 235.577039] ath11k_ce_recv_process_cb+0x15e/0x1e0 [ath11k]
[ 235.577041] ath11k_pci_ce_tasklet+0x10/0x30 [ath11k_pci]
[ 235.577043] tasklet_action_common.constprop.0+0xd4/0xf0
[ 235.577045] __do_softirq+0xc9/0x482
[ 235.577046] asm_call_on_stack+0x12/0x20
[ 235.577048] do_softirq_own_stack+0x49/0x60
[ 235.577049] irq_exit_rcu+0x9a/0xd0
[ 235.577050] common_interrupt+0xa1/0x190
[ 235.577052] asm_common_interrupt+0x1e/0x40
[ 235.577053] cpu_idle_poll.isra.0+0x2e/0x60
[ 235.577055] do_idle+0x5f/0xe0
[ 235.577056] cpu_startup_entry+0x14/0x20
[ 235.577058] start_kernel+0x443/0x464
[ 235.577060] secondary_startup_64+0xa4/0xb0
[ 235.577061] irq event stamp: 432035
[ 235.577063] hardirqs last enabled at (432035): [<ffffffff968d12b4>] _raw_spin_unlock_irqrestore+0x34/0x40
[ 235.577064] hardirqs last disabled at (432034): [<ffffffff968d10d3>] _raw_spin_lock_irqsave+0x63/0x80
[ 235.577066] softirqs last enabled at (431998): [<ffffffff967115c1>] inet6_fill_ifla6_attrs+0x3f1/0x430
[ 235.577067] softirqs last disabled at (431996): [<ffffffff9671159f>] inet6_fill_ifla6_attrs+0x3cf/0x430
[ 235.577068]
[ 235.577068] other info that might help us debug this:
[ 235.577069] Possible unsafe locking scenario:
[ 235.577069]
[ 235.577070] CPU0
[ 235.577070] ----
[ 235.577071] lock(&ab->base_lock);
[ 235.577072] <Interrupt>
[ 235.577073] lock(&ab->base_lock);
[ 235.577074]
[ 235.577074] *** DEADLOCK ***
[ 235.577074]
[ 235.577075] 3 locks held by kworker/u16:1/98:
[ 235.577076] #0: ffff9655f75b1d48 ((wq_completion)ath11k_qmi_driver_event){+.+.}-{0:0}, at: process_one_work+0x1d3/0x5d0
[ 235.577079] #1: ffffa33cc02f3e70 ((work_completion)(&ab->qmi.event_work)){+.+.}-{0:0}, at: process_one_work+0x1d3/0x5d0
[ 235.577081] #2: ffff9655f75cad50 (&ab->core_lock){+.+.}-{3:3}, at: ath11k_core_qmi_firmware_ready.part.0+0x4e/0x160 [ath11k]
[ 235.577087]
[ 235.577087] stack backtrace:
[ 235.577088] CPU: 3 PID: 98 Comm: kworker/u16:1 Not tainted 5.9.0-rc5-wt-ath+ #196
[ 235.577089] Hardware name: Intel(R) Client Systems NUC8i7HVK/NUC8i7HVB, BIOS HNKBLi70.86A.0049.2018.0801.1601 08/01/2018
[ 235.577095] Workqueue: ath11k_qmi_driver_event ath11k_qmi_driver_event_work [ath11k]
[ 235.577096] Call Trace:
[ 235.577100] dump_stack+0x77/0xa0
[ 235.577102] mark_lock_irq.cold+0x15/0x3c
[ 235.577104] mark_lock+0x1d7/0x540
[ 235.577105] mark_usage+0xc7/0x140
[ 235.577107] __lock_acquire+0x219/0x6e0
[ 235.577108] ? sched_clock_cpu+0xc/0xb0
[ 235.577110] lock_acquire+0xb6/0x270
[ 235.577116] ? ath11k_regd_update+0x28/0x1d0 [ath11k]
[ 235.577118] ? atomic_notifier_chain_register+0x2d/0x40
[ 235.577120] _raw_spin_lock+0x2c/0x70
[ 235.577125] ? ath11k_regd_update+0x28/0x1d0 [ath11k]
[ 235.577130] ath11k_regd_update+0x28/0x1d0 [ath11k]
[ 235.577136] __ath11k_mac_register+0x3fb/0x480 [ath11k]
[ 235.577141] ath11k_mac_register+0x119/0x180 [ath11k]
[ 235.577146] ath11k_core_pdev_create+0x17/0xe0 [ath11k]
[ 235.577150] ath11k_core_qmi_firmware_ready.part.0+0x65/0x160 [ath11k]
[ 235.577155] ath11k_qmi_driver_event_work+0x1c5/0x230 [ath11k]
[ 235.577158] process_one_work+0x265/0x5d0
[ 235.577160] worker_thread+0x49/0x300
[ 235.577161] ? process_one_work+0x5d0/0x5d0
[ 235.577163] kthread+0x135/0x150
[ 235.577164] ? kthread_create_worker_on_cpu+0x60/0x60
[ 235.577166] ret_from_fork+0x22/0x30Tested-on: QCA6390 hw2.0 PCI WLAN.HST.1.0.1-01740-QCAHSTSWPLZ_V2_TO_X86-1
--- drivers/net/wireless/ath/ath11k/reg.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/wireless/ath/ath11k/reg.c +3 -3
@@ -206,7 +206,7 @@ int ath11k_regd_update(struct ath11k *ar, bool init) ab = ar->ab; pdev_id = ar->pdev_idx; -spin_lock (&ab->base_lock); +spin_lock_bh (&ab->base_lock);if (init) {/* Apply the regd received during init through
@@ -227,7 +227,7 @@ int ath11k_regd_update(struct ath11k *ar, bool init)if (!regd) { ret = -EINVAL ; -spin_unlock (&ab->base_lock); +spin_unlock_bh (&ab->base_lock); goto err; }
@@ -238,7 +238,7 @@ int ath11k_regd_update(struct ath11k *ar, bool init)if (regd_copy)ath11k_copy_regd (regd, regd_copy); -spin_unlock (&ab->base_lock); +spin_unlock_bh (&ab->base_lock);if (!regd_copy) { ret = -ENOMEM ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 94/147 drivers/net/wan/hdlc_fr: Correctly handle special skb->protocol values
In reply to Sasha LevinFrom: Xie He <[email protected]>
[ Upstream commit 8306266c1d51aac9aa7aa907fe99032a58c6382c ]
The fr_hard_header function is used to prepend the header to skbs before transmission. It is used in 3 situations:
1) When a control packet is generated internally in this driver;
2) When a user sends an skb on an Ethernet-emulating PVC device;
3) When a user sends an skb on a normal PVC device.
These 3 situations need to be handled differently by fr_hard_header. Different headers should be prepended to the skb in different situations.
Currently fr_hard_header distinguishes these 3 situations using skb->protocol. For situation 1 and 2, a special skb->protocol value will be assigned before calling fr_hard_header, so that it can recognize these 2 situations. All skb->protocol values other than these special ones are treated by fr_hard_header as situation 3.
However, it is possible that in situation 3, the user sends an skb with one of the special skb->protocol values. In this case, fr_hard_header would incorrectly treat it as situation 1 or 2.
This patch tries to solve this issue by using skb->dev instead of skb->protocol to distinguish between these 3 situations. For situation 1, skb->dev would be NULL; for situation 2, skb->dev->type would be ARPHRD_ETHER; and for situation 3, skb->dev->type would be ARPHRD_DLCI.
This way fr_hard_header would be able to distinguish these 3 situations correctly regardless what skb->protocol value the user tries to use in situation 3.
--- drivers/net/wan/hdlc_fr.c | 98 ++++++++++++++++++++------------------- 1 file changed, 51 insertions(+), 47 deletions(-)
drivers/net/wan/hdlc_fr.c +51 -47
@@ -273,63 +273,69 @@ static inline struct net_device **get_dev_p(struct pvc_device *pvc,static int fr_hard_header (struct sk_buff **skb_p,u16 dlci) { -u16 head_len;struct sk_buff *skb = *skb_p; -switch (skb->protocol) { -case cpu_to_be16 (NLPID_CCITT_ANSI_LMI ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_CCITT_ANSI_LMI ; -break ; - -case cpu_to_be16 (NLPID_CISCO_LMI ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_CISCO_LMI ; -break ; - -case cpu_to_be16 (ETH_P_IP ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_IP ; -break ; - -case cpu_to_be16 (ETH_P_IPV6 ): - head_len =4 ; -skb_push (skb, head_len); - skb->data[3 ] =NLPID_IPV6 ; -break ; - -case cpu_to_be16 (ETH_P_802_3 ): - head_len =10 ; -if (skb_headroom (skb) < head_len) { -struct sk_buff *skb2 =skb_realloc_headroom (skb, - head_len); +if (!skb->dev) {/* Control packets */ +switch (dlci) { +case LMI_CCITT_ANSI_DLCI : +skb_push (skb,4 ); + skb->data[3 ] =NLPID_CCITT_ANSI_LMI ; +break ; + +case LMI_CISCO_DLCI : +skb_push (skb,4 ); + skb->data[3 ] =NLPID_CISCO_LMI ; +break ; + +default : +return -EINVAL ; + } + + }else if (skb->dev->type ==ARPHRD_DLCI ) { +switch (skb->protocol) { +case htons (ETH_P_IP ): +skb_push (skb,4 ); + skb->data[3 ] =NLPID_IP ; +break ; + +case htons (ETH_P_IPV6 ): +skb_push (skb,4 ); + skb->data[3 ] =NLPID_IPV6 ; +break ; + +default : +skb_push (skb,10 ); + skb->data[3 ] =FR_PAD ; + skb->data[4 ] =NLPID_SNAP ; +/* OUI 00-00-00 indicates an Ethertype follows */ + skb->data[5 ] =0x00 ; + skb->data[6 ] =0x00 ; + skb->data[7 ] =0x00 ; +/* This should be an Ethertype: */ + *(__be16 *)(skb->data +8 ) = skb->protocol; + } + + }else if (skb->dev->type ==ARPHRD_ETHER ) { +if (skb_headroom (skb) <10 ) { +struct sk_buff *skb2 =skb_realloc_headroom (skb,10 );if (!skb2)return -ENOBUFS ;dev_kfree_skb (skb); skb = *skb_p = skb2; } -skb_push (skb, head_len); +skb_push (skb,10 ); skb->data[3 ] =FR_PAD ; skb->data[4 ] =NLPID_SNAP ; - skb->data[5 ] =FR_PAD ; +/* OUI 00-80-C2 stands for the 802.1 organization */ + skb->data[5 ] =0x00 ; skb->data[6 ] =0x80 ; skb->data[7 ] =0xC2 ; +/* PID 00-07 stands for Ethernet frames without FCS */ skb->data[8 ] =0x00 ; - skb->data[9 ] =0x07 ;/* bridged Ethernet frame w/out FCS */ -break ; + skb->data[9 ] =0x07 ; -default : - head_len =10 ; -skb_push (skb, head_len); - skb->data[3 ] =FR_PAD ; - skb->data[4 ] =NLPID_SNAP ; - skb->data[5 ] =FR_PAD ; - skb->data[6 ] =FR_PAD ; - skb->data[7 ] =FR_PAD ; - *(__be16 *)(skb->data +8 ) = skb->protocol; + }else { +return -EINVAL ; }dlci_to_q922 (skb->data, dlci);
@@ -425,8 +431,8 @@ static netdev_tx_t pvc_xmit(struct sk_buff *skb, struct net_device *dev)skb_put (skb, pad);memset (skb->data + len,0 , pad); } - skb->protocol =cpu_to_be16 (ETH_P_802_3 ); } + skb->dev = dev;if (!fr_hard_header (&skb, pvc->dlci)) { dev->stats.tx_bytes += skb->len; dev->stats.tx_packets++;
@@ -494,10 +500,8 @@ static void fr_lmi_send(struct net_device *dev, int fullrep)memset (skb->data,0 , len);skb_reserve (skb,4 );if (lmi ==LMI_CISCO ) { - skb->protocol =cpu_to_be16 (NLPID_CISCO_LMI );fr_hard_header (&skb,LMI_CISCO_DLCI ); }else { - skb->protocol =cpu_to_be16 (NLPID_CCITT_ANSI_LMI );fr_hard_header (&skb,LMI_CCITT_ANSI_DLCI ); } data =skb_tail_pointer (skb);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 95/147 usb: dwc3: core: do not queue work if dr_mode is not USB_DR_MODE_OTG
In reply to Sasha LevinFrom: Li Jun <[email protected]>
[ Upstream commit dc336b19e82d0454ea60270cd18fbb4749e162f6 ]
Do not try to queue a drd work if dr_mode is not USB_DR_MODE_OTG because the work is not inited, this may be triggered by user try to change mode file of debugfs on a single role port, which will cause below kernel dump:
[ 60.115529] ------------[ cut here ]------------ [ 60.120166] WARNING: CPU: 1 PID: 627 at kernel/workqueue.c:1473 __queue_work+0x46c/0x520 [ 60.128254] Modules linked in: [ 60.131313] CPU: 1 PID: 627 Comm: sh Not tainted 5.7.0-rc4-00022-g914a586-dirty #135 [ 60.139054] Hardware name: NXP i.MX8MQ EVK (DT) [ 60.143585] pstate: a0000085 (NzCv daIf -PAN -UAO) [ 60.148376] pc : __queue_work+0x46c/0x520 [ 60.152385] lr : __queue_work+0x314/0x520 [ 60.156393] sp : ffff8000124ebc40 [ 60.159705] x29: ffff8000124ebc40 x28: ffff800011808018 [ 60.165018] x27: ffff800011819ef8 x26: ffff800011d39980 [ 60.170331] x25: ffff800011808018 x24: 0000000000000100 [ 60.175643] x23: 0000000000000013 x22: 0000000000000001 [ 60.180955] x21: ffff0000b7c08e00 x20: ffff0000b6c31080 [ 60.186267] x19: ffff0000bb99bc00 x18: 0000000000000000 [ 60.191579] x17: 0000000000000000 x16: 0000000000000000 [ 60.196891] x15: 0000000000000000 x14: 0000000000000000 [ 60.202202] x13: 0000000000000000 x12: 0000000000000000 [ 60.207515] x11: 0000000000000000 x10: 0000000000000040 [ 60.212827] x9 : ffff800011d55460 x8 : ffff800011d55458 [ 60.218138] x7 : ffff0000b7800028 x6 : 0000000000000000 [ 60.223450] x5 : ffff0000b7800000 x4 : 0000000000000000 [ 60.228762] x3 : ffff0000bb997cc0 x2 : 0000000000000001 [ 60.234074] x1 : 0000000000000000 x0 : ffff0000b6c31088 [ 60.239386] Call trace: [ 60.241834] __queue_work+0x46c/0x520 [ 60.245496] queue_work_on+0x6c/0x90 [ 60.249075] dwc3_set_mode+0x48/0x58 [ 60.252651] dwc3_mode_write+0xf8/0x150 [ 60.256489] full_proxy_write+0x5c/0xa8 [ 60.260327] __vfs_write+0x18/0x40 [ 60.263729] vfs_write+0xdc/0x1c8 [ 60.267045] ksys_write+0x68/0xf0 [ 60.270360] __arm64_sys_write+0x18/0x20 [ 60.274286] el0_svc_common.constprop.0+0x68/0x160 [ 60.279077] do_el0_svc+0x20/0x80 [ 60.282394] el0_sync_handler+0x10c/0x178 [ 60.286403] el0_sync+0x140/0x180 [ 60.289716] ---[ end trace 70b155582e2b7988 ]---
--- drivers/usb/dwc3/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/usb/dwc3/core.c +3 -3
@@ -120,9 +120,6 @@ static void __dwc3_set_mode(struct work_struct *work)unsigned long flags;int ret; -if (dwc->dr_mode !=USB_DR_MODE_OTG ) -return ; -pm_runtime_get_sync (dwc->dev);if (dwc->current_dr_role ==DWC3_GCTL_PRTCAP_OTG )
@@ -203,6 +200,9 @@ void dwc3_set_mode(struct dwc3 *dwc, u32 mode) {unsigned long flags; +if (dwc->dr_mode !=USB_DR_MODE_OTG ) +return ; +spin_lock_irqsave (&dwc->lock, flags); dwc->desired_dr_role = mode;spin_unlock_irqrestore (&dwc->lock, flags);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 96/147 bus: mhi: core: Abort suspends due to outgoing pending packets
In reply to Sasha LevinFrom: Bhaumik Bhatt <[email protected]>
[ Upstream commit 515847c557dd33167be86cb429fc0674a331bc88 ]
Add the missing check to abort suspends if a client driver has pending outgoing packets to send to the device. This allows better utilization of the MHI bus wherein clients on the host are not left waiting for longer suspend or resume cycles to finish for data transfers.
--- drivers/bus/mhi/core/pm.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/bus/mhi/core/pm.c +4 -2
@@ -686,7 +686,8 @@ int mhi_pm_suspend(struct mhi_controller *mhi_cntrl)return -EIO ;/* Return busy if there are any pending resources */ -if (atomic_read (&mhi_cntrl->dev_wake)) +if (atomic_read (&mhi_cntrl->dev_wake) || +atomic_read (&mhi_cntrl->pending_pkts))return -EBUSY ;/* Take MHI out of M2 state */
@@ -712,7 +713,8 @@ int mhi_pm_suspend(struct mhi_controller *mhi_cntrl)write_lock_irq (&mhi_cntrl->pm_lock); -if (atomic_read (&mhi_cntrl->dev_wake)) { +if (atomic_read (&mhi_cntrl->dev_wake) || +atomic_read (&mhi_cntrl->pending_pkts)) {write_unlock_irq (&mhi_cntrl->pm_lock);return -EBUSY ; }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 97/147 bus/fsl_mc: Do not rely on caller to provide non NULL mc_io
In reply to Sasha LevinFrom: Diana Craciun <[email protected]>
[ Upstream commit 5026cf605143e764e1785bbf9158559d17f8d260 ]
Before destroying the mc_io, check first that it was allocated.
--- drivers/bus/fsl-mc/mc-io.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/bus/fsl-mc/mc-io.c +6 -1
@@ -129,7 +129,12 @@ int __must_check fsl_create_mc_io(struct device *dev,*/ void fsl_destroy_mc_io (struct fsl_mc_io *mc_io) { -struct fsl_mc_device *dpmcp_dev = mc_io->dpmcp_dev; +struct fsl_mc_device *dpmcp_dev; + +if (!mc_io) +return ; + + dpmcp_dev = mc_io->dpmcp_dev;if (dpmcp_dev)fsl_mc_io_unset_dpmcp (mc_io);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 98/147 ACPI: HMAT: Fix handling of changes from ACPI 6.2 to ACPI 6.3
In reply to Sasha LevinFrom: Jonathan Cameron <[email protected]>
[ Upstream commit 2c5b9bde95c96942f2873cea6ef383c02800e4a8 ]
In ACPI 6.3, the Memory Proximity Domain Attributes Structure changed substantially. One of those changes was that the flag for "Memory Proximity Domain field is valid" was deprecated.
This was because the field "Proximity Domain for the Memory" became a required field and hence having a validity flag makes no sense.
So the correct logic is to always assume the field is there. Current code assumes it never is.
--- drivers/acpi/numa/hmat.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/acpi/numa/hmat.c +2 -1
@@ -424,7 +424,8 @@ static int __init hmat_parse_proximity_domain(union acpi_subtable_headers *headepr_info ("HMAT: Memory Flags:%04x Processor Domain:%u Memory Domain:%u\n" , p->flags, p->processor_PD, p->memory_PD); -if (p->flags &ACPI_HMAT_MEMORY_PD_VALID && hmat_revision ==1 ) { +if ((hmat_revision ==1 && p->flags &ACPI_HMAT_MEMORY_PD_VALID ) || + hmat_revision >1 ) { target =find_mem_target (p->memory_PD);if (!target) {pr_debug ("HMAT: Memory Domain missing from SRAT\n" );
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 99/147 power: supply: test_power: add missing newlines when printing parameters by sysfs
In reply to Sasha LevinFrom: Xiongfeng Wang <[email protected]>
[ Upstream commit c07fa6c1631333f02750cf59f22b615d768b4d8f ]
When I cat some module parameters by sysfs, it displays as follows. It's better to add a newline for easy reading.
root@syzkaller:~# cd /sys/module/test_power/parameters/ root@syzkaller:/sys/module/test_power/parameters# cat ac_online onroot@syzkaller:/sys/module/test_power/parameters# cat battery_present trueroot@syzkaller:/sys/module/test_power/parameters# cat battery_health goodroot@syzkaller:/sys/module/test_power/parameters# cat battery_status dischargingroot@syzkaller:/sys/module/test_power/parameters# cat battery_technology LIONroot@syzkaller:/sys/module/test_power/parameters# cat usb_online onroot@syzkaller:/sys/module/test_power/parameters#
--- drivers/power/supply/test_power.c | 6 ++++++ 1 file changed, 6 insertions(+)
drivers/power/supply/test_power.c +6 -0
@@ -353,6 +353,7 @@ static int param_set_ac_online(const char *key, const struct kernel_param *kp)static int param_get_ac_online (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_ac_online, ac_online,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -366,6 +367,7 @@ static int param_set_usb_online(const char *key, const struct kernel_param *kp)static int param_get_usb_online (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_ac_online, usb_online,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -380,6 +382,7 @@ static int param_set_battery_status(const char *key,static int param_get_battery_status (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_status, battery_status,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -394,6 +397,7 @@ static int param_set_battery_health(const char *key,static int param_get_battery_health (char *buffer,const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_health, battery_health,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -409,6 +413,7 @@ static int param_get_battery_present(char *buffer, const struct kernel_param *kp) {strcpy (buffer,map_get_key (map_present, battery_present,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
@@ -426,6 +431,7 @@ static int param_get_battery_technology(char *buffer, {strcpy (buffer,map_get_key (map_technology, battery_technology,"unknown" )); +strcat (buffer,"\n" );return strlen (buffer); }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 100/147 drm/amd/display: HDMI remote sink need mode validation for Linux
In reply to Sasha LevinFrom: Fangzhi Zuo <[email protected]>
[ Upstream commit 95d620adb48f7728e67d82f56f756e8d451cf8d2 ]
[Why]
Currently mode validation is bypassed if remote sink exists. That leads to mode set issue when a BW bottle neck exists in the link path, e.g., a DP-to-HDMI converter that only supports HDMI 1.4.
Any invalid mode passed to Linux user space will cause the modeset failure due to limitation of Linux user space implementation.
[How]
Mode validation is skipped only if in edid override. For real remote sink, clock limit check should be done for HDMI remote sink.
Have HDMI related remote sink going through mode validation to elimiate modes which pixel clock exceeds BW limitation.
--- drivers/gpu/drm/amd/display/dc/core/dc_link.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/display/dc/core/dc_link.c +1 -1
@@ -2441,7 +2441,7 @@ enum dc_status dc_link_validate_mode_timing(/* A hack to avoid failing any modes for EDID override feature on * topology change such as lower quality cable for DP or different dongle */ -if (link->remote_sinks[0 ]) +if (link->remote_sinks[0 ] && link->remote_sinks[0 ]->sink_signal ==SIGNAL_TYPE_VIRTUAL )return DC_OK ;/* Passive Dongle */
Signature · 6 lines
2.25.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 101/147 drm/amd/display: Avoid set zero in the requested clk
In reply to Sasha LevinFrom: Rodrigo Siqueira <[email protected]>
[ Upstream commit 2f8be0e516803cc3fd87c1671247896571a5a8fb ]
[Why]
Sometimes CRTCs can be disabled due to display unplugging or temporarily transition in the userspace; in these circumstances, DCE tries to set the minimum clock threshold. When we have this situation, the function bw_calcs is invoked with number_of_displays set to zero, making DCE set dispclk_khz and sclk_khz to zero. For these reasons, we have seen some ATOM bios errors that look like:
[drm:atom_op_jump [amdgpu]] *ERROR* atombios stuck in loop for more than 5secs aborting
[drm:amdgpu_atom_execute_table_locked [amdgpu]] *ERROR* atombios stuck executing EA8A (len 761, WS 0, PS 0) @ 0xEABA
[How]
This error happens due to an attempt to optimize the bandwidth using the sclk, and the dispclk clock set to zero. Technically we handle this in the function dce112_set_clock, but we are not considering the case that this value is set to zero. This commit fixes this issue by ensuring that we never set a minimum value below the minimum clock threshold.
--- drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/gpu/drm/amd/display/dc/clk_mgr/dce112/dce112_clk_mgr.c +1 -2
@@ -79,8 +79,7 @@ int dce112_set_clock(struct clk_mgr *clk_mgr_base, int requested_clk_khz)memset (&dce_clk_params,0 ,sizeof (dce_clk_params));/* Make sure requested clock isn't lower than minimum threshold*/ -if (requested_clk_khz >0 ) - requested_clk_khz =max (requested_clk_khz, + requested_clk_khz =max (requested_clk_khz, clk_mgr_dce->base.dentist_vco_freq_khz /62 ); dce_clk_params.target_clock_frequency = requested_clk_khz;
Signature · 6 lines
2.25.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 102/147 ARC: [dts] fix the errors detected by dtbs_check
In reply to Sasha LevinFrom: Zhen Lei <[email protected]>
[ Upstream commit 05b1be68c4d6d76970025e6139bfd735c2256ee5 ]
xxx/arc/boot/dts/axs101.dt.yaml: dw-apb-ictl@e0012000: $nodename:0: \
'dw-apb-ictl@e0012000' does not match '^interrupt-controller(@[0-9a-f,]+)*$'
From schema: xxx/interrupt-controller/snps,dw-apb-ictl.yaml
The node name of the interrupt controller must start with "interrupt-controller" instead of "dw-apb-ictl".
--- arch/arc/boot/dts/axc001.dtsi | 2 +- arch/arc/boot/dts/axc003.dtsi | 2 +- arch/arc/boot/dts/axc003_idu.dtsi | 2 +- arch/arc/boot/dts/vdk_axc003.dtsi | 2 +- arch/arc/boot/dts/vdk_axc003_idu.dtsi | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-)
arch/arc/boot/dts/axc001.dtsi +1 -1
@@ -91,7 +91,7 @@ arcpct0: pct { * avoid duplicating the MB dtsi file given that IRQ from * this intc to cpu intc are different for axs101 and axs103 */ - mb_intc: dw-apb-ictl@e0012000 { + mb_intc: interrupt-controller@e0012000 { #interrupt-cells = <1>; compatible = "snps,dw-apb-ictl"; reg = < 0x0 0xe0012000 0x0 0x200 >;
arch/arc/boot/dts/axc003.dtsi +1 -1
@@ -129,7 +129,7 @@ mmc@15000 { * avoid duplicating the MB dtsi file given that IRQ from * this intc to cpu intc are different for axs101 and axs103 */ - mb_intc: dw-apb-ictl@e0012000 { + mb_intc: interrupt-controller@e0012000 { #interrupt-cells = <1>; compatible = "snps,dw-apb-ictl"; reg = < 0x0 0xe0012000 0x0 0x200 >;
arch/arc/boot/dts/axc003_idu.dtsi +1 -1
@@ -135,7 +135,7 @@ mmc@15000 { * avoid duplicating the MB dtsi file given that IRQ from * this intc to cpu intc are different for axs101 and axs103 */ - mb_intc: dw-apb-ictl@e0012000 { + mb_intc: interrupt-controller@e0012000 { #interrupt-cells = <1>; compatible = "snps,dw-apb-ictl"; reg = < 0x0 0xe0012000 0x0 0x200 >;
arch/arc/boot/dts/vdk_axc003.dtsi +1 -1
@@ -46,7 +46,7 @@ debug_uart: dw-apb-uart@5000 { }; - mb_intc: dw-apb-ictl@e0012000 { + mb_intc: interrupt-controller@e0012000 { #interrupt-cells = <1>; compatible = "snps,dw-apb-ictl"; reg = < 0xe0012000 0x200 >;
arch/arc/boot/dts/vdk_axc003_idu.dtsi +1 -1
@@ -54,7 +54,7 @@ debug_uart: dw-apb-uart@5000 { }; - mb_intc: dw-apb-ictl@e0012000 { + mb_intc: interrupt-controller@e0012000 { #interrupt-cells = <1>; compatible = "snps,dw-apb-ictl"; reg = < 0xe0012000 0x200 >;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 103/147 block: Consider only dispatched requests for inflight statistic
In reply to Sasha LevinFrom: Gabriel Krisman Bertazi <[email protected]>
[ Upstream commit a926c7afffcc0f2e35e6acbccb16921bacf34617 ]
According to Documentation/block/stat.rst, inflight should not include I/O requests that are in the queue but not yet dispatched to the device, but blk-mq identifies as inflight any request that has a tag allocated, which, for queues without elevator, happens at request allocation time and before it is queued in the ctx (default case in blk_mq_submit_bio).
In addition, current behavior is different for queues with elevator from queues without it, since for the former the driver tag is allocated at dispatch time. A more precise approach would be to only consider requests with state MQ_RQ_IN_FLIGHT.
This effectively reverts commit 6131837b1de6 ("blk-mq: count allocated but not started requests in iostats inflight") to consolidate blk-mq behavior with itself (elevator case) and with original documentation, but it differs from the behavior used by the legacy path.
This version differs from v1 by using blk_mq_rq_state to access the state attribute. Avoid using blk_mq_request_started, which was suggested, since we don't want to include MQ_RQ_COMPLETE.
--- block/blk-mq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
block/blk-mq.c +1 -1
@@ -105,7 +105,7 @@ static bool blk_mq_check_inflight(struct blk_mq_hw_ctx *hctx, {struct mq_inflight *mi = priv; -if (rq->part == mi->part) +if (rq->part == mi->part &&blk_mq_rq_state (rq) ==MQ_RQ_IN_FLIGHT ) mi->inflight[rq_data_dir (rq)]++;return true;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 104/147 btrfs: fix replace of seed device
In reply to Sasha LevinFrom: Anand Jain <[email protected]>
[ Upstream commit c6a5d954950c5031444173ad2195efc163afcac9 ]
If you replace a seed device in a sprouted fs, it appears to have successfully replaced the seed device, but if you look closely, it didn't. Here is an example.
$ mkfs.btrfs /dev/sda $ btrfstune -S1 /dev/sda $ mount /dev/sda /btrfs $ btrfs device add /dev/sdb /btrfs $ umount /btrfs $ btrfs device scan --forget $ mount -o device=/dev/sda /dev/sdb /btrfs $ btrfs replace start -f /dev/sda /dev/sdc /btrfs $ echo $? 0
BTRFS info (device sdb): dev_replace from /dev/sda (devid 1) to /dev/sdc started BTRFS info (device sdb): dev_replace from /dev/sda (devid 1) to /dev/sdc finished
$ btrfs fi show Label: none uuid: ab2c88b7-be81-4a7e-9849-c3666e7f9f4f Total devices 2 FS bytes used 256.00KiB devid 1 size 3.00GiB used 520.00MiB path /dev/sdc devid 2 size 3.00GiB used 896.00MiB path /dev/sdb
Label: none uuid: 10bd3202-0415-43af-96a8-d5409f310a7e
Total devices 1 FS bytes used 128.00KiB
devid 1 size 3.00GiB used 536.00MiB path /dev/sda
So as per the replace start command and kernel log replace was successful. Now let's try to clean mount.
$ umount /btrfs $ btrfs device scan --forget
$ mount -o device=/dev/sdc /dev/sdb /btrfs mount: /btrfs: wrong fs type, bad option, bad superblock on /dev/sdb, missing codepage or helper program, or other error.
[ 636.157517] BTRFS error (device sdc): failed to read chunk tree: -2 [ 636.180177] BTRFS error (device sdc): open_ctree failed
That's because per dev items it is still looking for the original seed device.
$ btrfs inspect-internal dump-tree -d /dev/sdb
item 0 key (DEV_ITEMS DEV_ITEM 1) itemoff 16185 itemsize 98 devid 1 total_bytes 3221225472 bytes_used 545259520 io_align 4096 io_width 4096 sector_size 4096 type 0 generation 6 start_offset 0 dev_group 0 seek_speed 0 bandwidth 0 uuid 59368f50-9af2-4b17-91da-8a783cc418d4 <--- seed uuid fsid 10bd3202-0415-43af-96a8-d5409f310a7e <--- seed fsid item 1 key (DEV_ITEMS DEV_ITEM 2) itemoff 16087 itemsize 98 devid 2 total_bytes 3221225472 bytes_used 939524096 io_align 4096 io_width 4096 sector_size 4096 type 0 generation 0 start_offset 0 dev_group 0 seek_speed 0 bandwidth 0 uuid 56a0a6bc-4630-4998-8daf-3c3030c4256a <- sprout uuid fsid ab2c88b7-be81-4a7e-9849-c3666e7f9f4f <- sprout fsid
But the replaced target has the following uuid+fsid in its superblock which doesn't match with the expected uuid+fsid in its devitem.
$ btrfs in dump-super /dev/sdc | egrep '^generation|dev_item.uuid|dev_item.fsid|devid' generation 20 dev_item.uuid 59368f50-9af2-4b17-91da-8a783cc418d4 dev_item.fsid ab2c88b7-be81-4a7e-9849-c3666e7f9f4f [match] dev_item.devid 1
So if you provide the original seed device the mount shall be successful. Which so long happening in the test case btrfs/163.
$ btrfs device scan --forget $ mount -o device=/dev/sda /dev/sdb /btrfs
Fix in this patch:
If a seed is not sprouted then there is no replacement of it, because of its read-only filesystem with a read-only device. Similarly, in the case of a sprouted filesystem, the seed device is still read only. So, mark it as you can't replace a seed device, you can only add a new device and then delete the seed device. If replace is attempted then returns -EINVAL.
--- fs/btrfs/dev-replace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/dev-replace.c +1 -1
@@ -230,7 +230,7 @@ static int btrfs_init_dev_replace_tgtdev(struct btrfs_fs_info *fs_info,int ret =0 ; *device_out =NULL ; -if (fs_info->fs_devices->seeding) { +if (srcdev->fs_devices->seeding) {btrfs_err (fs_info,"the filesystem is a seed filesystem!" );return -EINVAL ; }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 105/147 md/bitmap: md_bitmap_get_counter returns wrong blocks
In reply to Sasha LevinFrom: Zhao Heming <[email protected]>
[ Upstream commit d837f7277f56e70d82b3a4a037d744854e62f387 ]
md_bitmap_get_counter() has code:
```
if (bitmap->bp[page].hijacked ||
bitmap->bp[page].map == NULL)
csize = ((sector_t)1) << (bitmap->chunkshift +
PAGE_COUNTER_SHIFT - 1);
```The minus 1 is wrong, this branch should report 2048 bits of space. With "-1" action, this only report 1024 bit of space.
This bug code returns wrong blocks, but it doesn't inflence bitmap logic:
1. Most callers focus this function return value (the counter of offset), not the parameter blocks.
2. The bug is only triggered when hijacked is true or map is NULL. the hijacked true condition is very rare.
the "map == null" only true when array is creating or resizing.
3. Even the caller gets wrong blocks, current code makes caller just to call md_bitmap_get_counter() one more time.
--- drivers/md/md-bitmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/md/md-bitmap.c +1 -1
@@ -1367,7 +1367,7 @@ __acquires(bitmap->lock)if (bitmap->bp[page].hijacked || bitmap->bp[page].map ==NULL ) csize = ((sector_t )1 ) << (bitmap->chunkshift + -PAGE_COUNTER_SHIFT -1 ); +PAGE_COUNTER_SHIFT );else csize = ((sector_t )1 ) << bitmap->chunkshift; *blocks = csize - (offset & (csize -1 ));
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 106/147 f2fs: fix to set SBI_NEED_FSCK flag for inconsistent inode
In reply to Sasha LevinFrom: Chao Yu <[email protected]>
[ Upstream commit d662fad143c0470ad7f46ea7b02da539f613d7d7 ]
If compressed inode has inconsistent fields on i_compress_algorithm, i_compr_blocks and i_log_cluster_size, we missed to set SBI_NEED_FSCK to notice fsck to repair the inode, fix it.
--- fs/f2fs/inode.c | 3 +++ 1 file changed, 3 insertions(+)
fs/f2fs/inode.c +3 -0
@@ -292,6 +292,7 @@ static bool sanity_check_inode(struct inode *inode, struct page *node_page) F2FS_FITS_IN_INODE(ri, fi->i_extra_isize, i_log_cluster_size)) {if (ri->i_compress_algorithm >=COMPRESS_MAX ) { +set_sbi_flag (sbi,SBI_NEED_FSCK ); f2fs_warn(sbi, "%s: inode (ino=%lx) has unsupported " "compress algorithm: %u, run fsck to fix", __func__, inode->i_ino,
@@ -300,6 +301,7 @@ static bool sanity_check_inode(struct inode *inode, struct page *node_page) }if (le64_to_cpu (ri->i_compr_blocks) >SECTOR_TO_BLOCK (inode->i_blocks)) { +set_sbi_flag (sbi,SBI_NEED_FSCK ); f2fs_warn(sbi, "%s: inode (ino=%lx) has inconsistent " "i_compr_blocks:%llu, i_blocks:%llu, run fsck to fix", __func__, inode->i_ino,
@@ -309,6 +311,7 @@ static bool sanity_check_inode(struct inode *inode, struct page *node_page) }if (ri->i_log_cluster_size <MIN_COMPRESS_LOG_SIZE || ri->i_log_cluster_size >MAX_COMPRESS_LOG_SIZE ) { +set_sbi_flag (sbi,SBI_NEED_FSCK ); f2fs_warn(sbi, "%s: inode (ino=%lx) has unsupported " "log cluster size: %u, run fsck to fix", __func__, inode->i_ino,
Signature · 8 lines
2.25.1 _______________________________________________ Linux-f2fs-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Sasha Levin
maintainerPATCH 107/147 bnxt_en: Log unknown link speed appropriately.
In reply to Sasha LevinFrom: Michael Chan <[email protected]>
[ Upstream commit 8eddb3e7ce124dd6375d3664f1aae13873318b0f ]
If the VF virtual link is set to always enabled, the speed may be unknown when the physical link is down. The driver currently logs the link speed as 4294967295 Mbps which is SPEED_UNKNOWN. Modify the link up log message as "speed unknown" which makes more sense.
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +5 -1
@@ -8744,6 +8744,11 @@ static void bnxt_report_link(struct bnxt *bp)u16 fec;netif_carrier_on (bp->dev); + speed =bnxt_fw_to_ethtool_speed (bp->link_info.link_speed); +if (speed ==SPEED_UNKNOWN ) { +netdev_info (bp->dev,"NIC Link is Up, speed unknown\n" ); +return ; + }if (bp->link_info.duplex ==BNXT_LINK_DUPLEX_FULL ) duplex ="full" ;else
@@ -8756,7 +8761,6 @@ static void bnxt_report_link(struct bnxt *bp) flow_ctrl ="ON - receive" ; else flow_ctrl ="none" ; - speed =bnxt_fw_to_ethtool_speed (bp->link_info.link_speed);netdev_info (bp->dev,"NIC Link is Up, %u Mbps %s duplex, Flow control: %s\n" , speed, duplex, flow_ctrl);if (bp->flags &BNXT_FLAG_EEE_CAP )
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 108/147 binfmt_elf: take the mmap lock around find_extend_vma()
In reply to Sasha LevinFrom: Jann Horn <[email protected]>
[ Upstream commit b2767d97f5ff758250cf28684aaa48bbfd34145f ]
create_elf_tables() runs after setup_new_exec(), so other tasks can already access our new mm and do things like process_madvise() on it. (At the time I'm writing this commit, process_madvise() is not in mainline yet, but has been in akpm's tree for some time.)
While I believe that there are currently no APIs that would actually allow another process to mess up our VMA tree (process_madvise() is limited to MADV_COLD and MADV_PAGEOUT, and uring and userfaultfd cannot reach an mm under which no syscalls have been executed yet), this seems like an accident waiting to happen.
Let's make sure that we always take the mmap lock around GUP paths as long as another process might be able to see the mm.
(Yes, this diff looks suspicious because we drop the lock before doing anything with vma, but that's because we actually don't do anything with it apart from the NULL check.)
--- fs/binfmt_elf.c | 3 +++ 1 file changed, 3 insertions(+)
fs/binfmt_elf.c +3 -0
@@ -309,7 +309,10 @@ create_elf_tables(struct linux_binprm *bprm, const struct elfhdr *exec,* Grow the stack manually; some architectures have a limit on how * far ahead a user-space access may be in order to grow the stack. */ +if (mmap_read_lock_killable (mm)) +return -EINTR ; vma =find_extend_vma (mm, bprm->p); +mmap_read_unlock (mm);if (!vma)return -EFAULT ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 109/147 rpmsg: glink: Use complete_all for open states
In reply to Sasha LevinFrom: Chris Lew <[email protected]>
[ Upstream commit 4fcdaf6e28d11e2f3820d54dd23cd12a47ddd44e ]
The open_req and open_ack completion variables are the state variables to represet a remote channel as open. Use complete_all so there are no races with waiters and using completion_done.
--- drivers/rpmsg/qcom_glink_native.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/rpmsg/qcom_glink_native.c +3 -3
@@ -970,7 +970,7 @@ static int qcom_glink_rx_open_ack(struct qcom_glink *glink, unsigned int lcid)return -EINVAL ; } -complete (&channel->open_ack); +complete_all (&channel->open_ack);return 0 ; }
@@ -1178,7 +1178,7 @@ static int qcom_glink_announce_create(struct rpmsg_device *rpdev)__be32 *val = defaults;int size; -if (glink->intentless) +if (glink->intentless || !completion_done (&channel->open_ack))return 0 ; prop =of_find_property (np,"qcom,intents" ,NULL );
@@ -1413,7 +1413,7 @@ static int qcom_glink_rx_open(struct qcom_glink *glink, unsigned int rcid, channel->rcid = ret;spin_unlock_irqrestore (&glink->idr_lock, flags); -complete (&channel->open_req); +complete_all (&channel->open_req);if (create_device) { rpdev =kzalloc (sizeof (*rpdev),GFP_KERNEL );
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 110/147 PCI/ACPI: Add Ampere Altra SOC MCFG quirk
In reply to Sasha LevinFrom: Tuan Phan <[email protected]>
[ Upstream commit 877c1a5f79c6984bbe3f2924234c08e2f4f1acd5 ]
Ampere Altra SOC supports only 32-bit ECAM reads. Add an MCFG quirk for the platform.
--- drivers/acpi/pci_mcfg.c | 20 ++++++++++++++++++++ drivers/pci/ecam.c | 10 ++++++++++ include/linux/pci-ecam.h | 1 + 3 files changed, 31 insertions(+)
drivers/acpi/pci_mcfg.c +20 -0
@@ -142,6 +142,26 @@ static struct mcfg_fixup mcfg_quirks[] = {XGENE_V2_ECAM_MCFG (4 ,0 ),XGENE_V2_ECAM_MCFG (4 ,1 ),XGENE_V2_ECAM_MCFG (4 ,2 ), + +#define ALTRA_ECAM_QUIRK(rev, seg) \ + { "Ampere", "Altra ", rev, seg, MCFG_BUS_ANY, &pci_32b_read_ops } + +ALTRA_ECAM_QUIRK (1 ,0 ), +ALTRA_ECAM_QUIRK (1 ,1 ), +ALTRA_ECAM_QUIRK (1 ,2 ), +ALTRA_ECAM_QUIRK (1 ,3 ), +ALTRA_ECAM_QUIRK (1 ,4 ), +ALTRA_ECAM_QUIRK (1 ,5 ), +ALTRA_ECAM_QUIRK (1 ,6 ), +ALTRA_ECAM_QUIRK (1 ,7 ), +ALTRA_ECAM_QUIRK (1 ,8 ), +ALTRA_ECAM_QUIRK (1 ,9 ), +ALTRA_ECAM_QUIRK (1 ,10 ), +ALTRA_ECAM_QUIRK (1 ,11 ), +ALTRA_ECAM_QUIRK (1 ,12 ), +ALTRA_ECAM_QUIRK (1 ,13 ), +ALTRA_ECAM_QUIRK (1 ,14 ), + ALTRA_ECAM_QUIRK(1, 15), };static char mcfg_oem_id[ACPI_OEM_ID_SIZE ];
drivers/pci/ecam.c +10 -0
@@ -168,4 +168,14 @@ const struct pci_ecam_ops pci_32b_ops = { .write = pci_generic_config_write32, } }; + +/* ECAM ops for 32-bit read only (non-compliant) */ +const struct pci_ecam_ops pci_32b_read_ops = { + .bus_shift =20 , + .pci_ops = { + .map_bus = pci_ecam_map_bus, + .read = pci_generic_config_read32, + .write = pci_generic_config_write, + } +};#endif
include/linux/pci-ecam.h +1 -0
@@ -51,6 +51,7 @@ extern const struct pci_ecam_ops pci_generic_ecam_ops;#if defined(CONFIG_ACPI ) && defined(CONFIG_PCI_QUIRKS )extern const struct pci_ecam_ops pci_32b_ops;/* 32-bit accesses only */ +extern const struct pci_ecam_ops pci_32b_read_ops;/* 32-bit read only */ extern const struct pci_ecam_ops hisi_pcie_ops;/* HiSilicon */ extern const struct pci_ecam_ops thunder_pem_ecam_ops;/* Cavium ThunderX 1.x & 2.x */ extern const struct pci_ecam_ops pci_thunder_ecam_ops;/* Cavium ThunderX 1.x */
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 111/147 clk: ti: clockdomain: fix static checker warning
In reply to Sasha LevinFrom: Tero Kristo <[email protected]>
[ Upstream commit b7a7943fe291b983b104bcbd2f16e8e896f56590 ]
Fix a memory leak induced by not calling clk_put after doing of_clk_get.
--- drivers/clk/ti/clockdomain.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/clk/ti/clockdomain.c +2 -0
@@ -148,10 +148,12 @@ static void __init of_ti_clockdomain_setup(struct device_node *node)if (!omap2_clk_is_hw_omap (clk_hw)) {pr_warn ("can't setup clkdm for basic clk %s\n" ,__clk_get_name (clk)); +clk_put (clk);continue ; }to_clk_hw_omap (clk_hw)->clkdm_name = clkdm_name;omap2_init_clk_clkdm (clk_hw); +clk_put (clk); } }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 112/147 nfsd: rename delegation related tracepoints to make them less confusing
In reply to Sasha LevinFrom: Hou Tao <[email protected]>
[ Upstream commit 3caf91757ced158e6c4a44d8b105bd7b3e1767d8 ]
Now when a read delegation is given, two delegation related traces will be printed:
nfsd_deleg_open: client 5f45b854:e6058001 stateid 00000030:00000001
nfsd_deleg_none: client 5f45b854:e6058001 stateid 0000002f:00000001Although the intention is to let developers know two stateid are returned, the traces are confusing about whether or not a read delegation is handled out. So renaming trace_nfsd_deleg_none() to trace_nfsd_open() and trace_nfsd_deleg_open() to trace_nfsd_deleg_read() to make the intension clearer.
The patched traces will be:
nfsd_deleg_read: client 5f48a967:b55b21cd stateid 00000003:00000001
nfsd_open: client 5f48a967:b55b21cd stateid 00000002:00000001--- fs/nfsd/nfs4state.c | 4 ++-- fs/nfsd/trace.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-)
fs/nfsd/nfs4state.c +2 -2
@@ -5126,7 +5126,7 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open,memcpy (&open->op_delegate_stateid, &dp->dl_stid.sc_stateid,sizeof (dp->dl_stid.sc_stateid)); -trace_nfsd_deleg_open (&dp->dl_stid.sc_stateid); +trace_nfsd_deleg_read (&dp->dl_stid.sc_stateid); open->op_delegate_type =NFS4_OPEN_DELEGATE_READ ;nfs4_put_stid (&dp->dl_stid);return ;
@@ -5243,7 +5243,7 @@ nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfnfs4_open_delegation (current_fh, open, stp); nodeleg: status = nfs_ok; -trace_nfsd_deleg_none (&stp->st_stid.sc_stateid); +trace_nfsd_open (&stp->st_stid.sc_stateid); out:/* 4.1 client trying to upgrade/downgrade delegation? */ if (open->op_delegate_type == NFS4_OPEN_DELEGATE_NONE && dp &&
fs/nfsd/trace.h +2 -2
@@ -289,8 +289,8 @@ DEFINE_STATEID_EVENT(layout_recall_done);DEFINE_STATEID_EVENT (layout_recall_fail);DEFINE_STATEID_EVENT (layout_recall_release); -DEFINE_STATEID_EVENT (deleg_open); -DEFINE_STATEID_EVENT (deleg_none); +DEFINE_STATEID_EVENT (open); +DEFINE_STATEID_EVENT (deleg_read);DEFINE_STATEID_EVENT (deleg_break);DEFINE_STATEID_EVENT (deleg_recall);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 113/147 nfsd4: remove check_conflicting_opens warning
In reply to Sasha LevinFrom: "J. Bruce Fields" <[email protected]>
[ Upstream commit 50747dd5e47bde3b7d7f839c84d0d3b554090497 ]
There are actually rare races where this is possible (e.g. if a new open intervenes between the read of i_writecount and the fi_fds).
--- fs/nfsd/nfs4state.c | 1 - 1 file changed, 1 deletion(-)
fs/nfsd/nfs4state.c +0 -1
@@ -4954,7 +4954,6 @@ static int nfsd4_check_conflicting_opens(struct nfs4_client *clp, writes--;if (fp->fi_fds[O_RDWR ]) writes--; -WARN_ON_ONCE (writes <0 );if (writes >0 )return -EAGAIN ;spin_lock (&fp->fi_lock);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 114/147 phy: marvell: comphy: Convert internal SMCC firmware return codes to errno
In reply to Sasha LevinFrom: Pali Rohár <[email protected]>
[ Upstream commit ea17a0f153af2cd890e4ce517130dcccaa428c13 ]
Driver ->power_on and ->power_off callbacks leaks internal SMCC firmware return codes to phy caller. This patch converts SMCC error codes to standard linux errno codes. Include file linux/arm-smccc.h already provides defines for SMCC error codes, so use them instead of custom driver defines. Note that return value is signed 32bit, but stored in unsigned long type with zero padding.
--- drivers/phy/marvell/phy-mvebu-a3700-comphy.c | 14 +++++++++++--- drivers/phy/marvell/phy-mvebu-cp110-comphy.c | 14 +++++++++++--- 2 files changed, 22 insertions(+), 6 deletions(-)
drivers/phy/marvell/phy-mvebu-a3700-comphy.c +11 -3
@@ -26,7 +26,6 @@#define COMPHY_SIP_POWER_ON 0x82000001#define COMPHY_SIP_POWER_OFF 0x82000002#define COMPHY_SIP_PLL_LOCK 0x82000003 -#define COMPHY_FW_NOT_SUPPORTED (-1)#define COMPHY_FW_MODE_SATA 0x1#define COMPHY_FW_MODE_SGMII 0x2
@@ -112,10 +111,19 @@ static int mvebu_a3700_comphy_smc(unsigned long function, unsigned long lane, unsigned long mode) {struct arm_smccc_res res; +s32 ret;arm_smccc_smc (function, lane, mode,0 ,0 ,0 ,0 ,0 , &res); + ret = res.a0; -return res.a0; +switch (ret) { +case SMCCC_RET_SUCCESS : +return 0 ; +case SMCCC_RET_NOT_SUPPORTED : +return -EOPNOTSUPP ; +default : +return -EINVAL ; + } } static int mvebu_a3700_comphy_get_fw_mode(int lane, int port,
@@ -220,7 +228,7 @@ static int mvebu_a3700_comphy_power_on(struct phy *phy) } ret =mvebu_a3700_comphy_smc (COMPHY_SIP_POWER_ON , lane->id, fw_param); -if (ret ==COMPHY_FW_NOT_SUPPORTED ) +if (ret == -EOPNOTSUPP )dev_err (lane->dev,"unsupported SMC call, try updating your firmware\n" );
drivers/phy/marvell/phy-mvebu-cp110-comphy.c +11 -3
@@ -123,7 +123,6 @@#define COMPHY_SIP_POWER_ON 0x82000001#define COMPHY_SIP_POWER_OFF 0x82000002 -#define COMPHY_FW_NOT_SUPPORTED (-1)/* * A lane is described by the following bitfields:
@@ -273,10 +272,19 @@ static int mvebu_comphy_smc(unsigned long function, unsigned long phys, unsigned long lane, unsigned long mode) {struct arm_smccc_res res; +s32 ret;arm_smccc_smc (function, phys, lane, mode,0 ,0 ,0 ,0 , &res); + ret = res.a0; -return res.a0; +switch (ret) { +case SMCCC_RET_SUCCESS : +return 0 ; +case SMCCC_RET_NOT_SUPPORTED : +return -EOPNOTSUPP ; +default : +return -EINVAL ; + } } static int mvebu_comphy_get_mode(bool fw_mode, int lane, int port,
@@ -819,7 +827,7 @@ static int mvebu_comphy_power_on(struct phy *phy)if (!ret)return ret; -if (ret ==COMPHY_FW_NOT_SUPPORTED ) +if (ret == -EOPNOTSUPP )dev_err (priv->dev,"unsupported SMC call, try updating your firmware\n" );
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 115/147 net: 9p: initialize sun_server.sun_path to have addr's value only when addr is valid
In reply to Sasha LevinFrom: Anant Thazhemadam <[email protected]>
[ Upstream commit 7ca1db21ef8e0e6725b4d25deed1ca196f7efb28 ]
In p9_fd_create_unix, checking is performed to see if the addr (passed as an argument) is NULL or not.
However, no check is performed to see if addr is a valid address, i.e., it doesn't entirely consist of only 0's.
The initialization of sun_server.sun_path to be equal to this faulty addr value leads to an uninitialized variable, as detected by KMSAN. Checking for this (faulty addr) and returning a negative error number appropriately, resolves this issue.
--- net/9p/trans_fd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/9p/trans_fd.c +1 -1
@@ -1023,7 +1023,7 @@ p9_fd_create_unix(struct p9_client *client, const char *addr, char *args) csocket =NULL ; -if (addr ==NULL ) +if (!addr || !strlen (addr))return -EINVAL ;if (strlen (addr) >=UNIX_PATH_MAX ) {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 116/147 ceph: encode inodes' parent/d_name in cap reconnect message
In reply to Sasha LevinFrom: "Yan, Zheng" <[email protected]>
[ Upstream commit a33f6432b3a63a4909dbbb0967f7c9df8ff2de91 ]
Since nautilus, MDS tracks dirfrags whose child inodes have caps in open file table. When MDS recovers, it prefetches all of these dirfrags. This avoids using backtrace to load inodes. But dirfrags prefetch may load lots of useless inodes into cache, and make MDS run out of memory.
Recent MDS adds an option that disables dirfrags prefetch. When dirfrags prefetch is disabled. Recovering MDS only prefetches corresponding dir inodes. Including inodes' parent/d_name in cap reconnect message can help MDS to load inodes into its cache.
--- fs/ceph/mds_client.c | 89 ++++++++++++++++++++++++++++++-------------- 1 file changed, 61 insertions(+), 28 deletions(-)
fs/ceph/mds_client.c +61 -28
@@ -3612,6 +3612,39 @@ static int send_reconnect_partial(struct ceph_reconnect_state *recon_state)return err; } +static struct dentry *d_find_primary (struct inode *inode) +{ +struct dentry *alias, *dn =NULL ; + +if (hlist_empty (&inode->i_dentry)) +return NULL ; + +spin_lock (&inode->i_lock); +if (hlist_empty (&inode->i_dentry)) + goto out_unlock; + +if (S_ISDIR (inode->i_mode)) { + alias = hlist_entry(inode->i_dentry.first, struct dentry, d_u.d_alias); +if (!IS_ROOT (alias)) + dn =dget (alias); + goto out_unlock; + } + +hlist_for_each_entry (alias, &inode->i_dentry, d_u.d_alias) { +spin_lock (&alias->d_lock); +if (!d_unhashed (alias) && + (ceph_dentry (alias)->flags &CEPH_DENTRY_PRIMARY_LINK )) { + dn =dget_dlock (alias); + } +spin_unlock (&alias->d_lock); +if (dn) +break ; + } +out_unlock: +spin_unlock (&inode->i_lock); +return dn; +} +/* * Encode information about a cap for a reconnect with the MDS. */
@@ -3625,13 +3658,32 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap,struct ceph_inode_info *ci = cap->ci;struct ceph_reconnect_state *recon_state = arg;struct ceph_pagelist *pagelist = recon_state->pagelist; -int err; +struct dentry *dentry; +char *path; +int pathlen, err; +u64 pathbase;u64 snap_follows;dout (" adding %p ino %llx.%llx cap %p %lld %s\n" , inode,ceph_vinop (inode), cap, cap->cap_id,ceph_cap_string (cap->issued)); + dentry =d_find_primary (inode); +if (dentry) { +/* set pathbase to parent dir when msg_version >= 2 */ + path =ceph_mdsc_build_path (dentry, &pathlen, &pathbase, + recon_state->msg_version >=2 ); +dput (dentry); +if (IS_ERR (path)) { + err =PTR_ERR (path); + goto out_err; + } + }else { + path =NULL ; + pathlen =0 ; + pathbase =0 ; + } +spin_lock (&ci->i_ceph_lock); cap->seq =0 ;/* reset cap seq */ cap->issue_seq =0 ;/* and issue_seq */
@@ -3652,7 +3704,7 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap, rec.v2.wanted =cpu_to_le32 (__ceph_caps_wanted (ci)); rec.v2.issued =cpu_to_le32 (cap->issued); rec.v2.snaprealm =cpu_to_le64 (ci->i_snap_realm->ino); - rec.v2.pathbase =0 ; + rec.v2.pathbase =cpu_to_le64 (pathbase); rec.v2.flock_len = (__force __le32) ((ci->i_ceph_flags &CEPH_I_ERROR_FILELOCK ) ?0 :1 ); }else {
@@ -3663,7 +3715,7 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap,ceph_encode_timespec64 (&rec.v1.mtime, &inode->i_mtime);ceph_encode_timespec64 (&rec.v1.atime, &inode->i_atime); rec.v1.snaprealm =cpu_to_le64 (ci->i_snap_realm->ino); - rec.v1.pathbase =0 ; + rec.v1.pathbase =cpu_to_le64 (pathbase); }if (list_empty (&ci->i_cap_snaps)) {
@@ -3725,7 +3777,7 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap,sizeof (struct ceph_filelock ); rec.v2.flock_len =cpu_to_le32 (struct_len); - struct_len +=sizeof (u32) +sizeof (rec.v2); + struct_len +=sizeof (u32) + pathlen +sizeof (rec.v2);if (struct_v >=2 ) struct_len +=sizeof (u64);/* snap_follows */
@@ -3749,7 +3801,7 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap,ceph_pagelist_encode_8 (pagelist,1 );ceph_pagelist_encode_32 (pagelist, struct_len); } -ceph_pagelist_encode_string (pagelist,NULL ,0 ); +ceph_pagelist_encode_string (pagelist, path, pathlen);ceph_pagelist_append (pagelist, &rec,sizeof (rec.v2));ceph_locks_to_pagelist (flocks, pagelist, num_fcntl_locks, num_flock_locks);
@@ -3758,39 +3810,20 @@ static int reconnect_caps_cb(struct inode *inode, struct ceph_cap *cap, out_freeflocks:kfree (flocks); }else { -u64 pathbase =0 ; -int pathlen =0 ; -char *path =NULL ; -struct dentry *dentry; - - dentry =d_find_alias (inode); -if (dentry) { - path =ceph_mdsc_build_path (dentry, - &pathlen, &pathbase,0 ); -dput (dentry); -if (IS_ERR (path)) { - err =PTR_ERR (path); - goto out_err; - } - rec.v1.pathbase =cpu_to_le64 (pathbase); - } - err =ceph_pagelist_reserve (pagelist,sizeof (u64) +sizeof (u32) + pathlen +sizeof (rec.v1)); -if (err) { - goto out_freepath; - } +if (err) + goto out_err;ceph_pagelist_encode_64 (pagelist,ceph_ino (inode));ceph_pagelist_encode_string (pagelist, path, pathlen);ceph_pagelist_append (pagelist, &rec,sizeof (rec.v1)); -out_freepath: -ceph_mdsc_free_path (path, pathlen); } out_err: -if (err >=0 ) +ceph_mdsc_free_path (path, pathlen); +if (!err) recon_state->nr_caps++;return err; }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 117/147 drivers: watchdog: rdc321x_wdt: Fix race condition bugs
In reply to Sasha LevinFrom: Madhuparna Bhowmik <[email protected]>
[ Upstream commit 4b2e7f99cdd314263c9d172bc17193b8b6bba463 ]
In rdc321x_wdt_probe(), rdc321x_wdt_device.queue is initialized after misc_register(), hence if ioctl is called before its initialization which can call rdc321x_wdt_start() function, it will see an uninitialized value of rdc321x_wdt_device.queue, hence initialize it before misc_register().
Also, rdc321x_wdt_device.default_ticks is accessed in reset() function called from write callback, thus initialize it before misc_register().
Found by Linux Driver Verification project (linuxtesting.org).
--- drivers/watchdog/rdc321x_wdt.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/watchdog/rdc321x_wdt.c +2 -3
@@ -231,6 +231,8 @@ static int rdc321x_wdt_probe(struct platform_device *pdev) rdc321x_wdt_device.sb_pdev = pdata->sb_pdev; rdc321x_wdt_device.base_reg = r->start; + rdc321x_wdt_device.queue =0 ; + rdc321x_wdt_device.default_ticks = ticks; err =misc_register (&rdc321x_wdt_misc);if (err <0 ) {
@@ -245,14 +247,11 @@ static int rdc321x_wdt_probe(struct platform_device *pdev) rdc321x_wdt_device.base_reg, RDC_WDT_RST);init_completion (&rdc321x_wdt_device.stop); - rdc321x_wdt_device.queue =0 ;clear_bit (0 , &rdc321x_wdt_device.inuse);timer_setup (&rdc321x_wdt_device.timer, rdc321x_wdt_trigger,0 ); - rdc321x_wdt_device.default_ticks = ticks; -dev_info (&pdev->dev,"watchdog init success\n" );return 0 ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 118/147 jbd2: avoid transaction reuse after reformatting
In reply to Sasha LevinFrom: changfengnan <[email protected]>
[ Upstream commit fc750a3b44bdccb9fb96d6abbc48a9b8e480ce7b ]
When ext4 is formatted with lazy_journal_init=1 and transactions from the previous filesystem are still on disk, it is possible that they are considered during a recovery after a crash. Because the checksum seed has changed, the CRC check will fail, and the journal recovery fails with checksum error although the journal is otherwise perfectly valid. Fix the problem by checking commit block time stamps to determine whether the data in the journal block is just stale or whether it is indeed corrupt.
--- fs/jbd2/recovery.c | 78 +++++++++++++++++++++++++++++++++++++++------- 1 file changed, 66 insertions(+), 12 deletions(-)
fs/jbd2/recovery.c +66 -12
@@ -428,6 +428,8 @@ static int do_one_pass(journal_t *journal,__u32 crc32_sum = ~0 ;/* Transactional Checksums */ int descr_csum_size =0 ;int block_error =0 ; +bool need_check_commit_time = false; +__u64 last_trans_commit_time =0 , commit_time;/* * First thing is to establish what we expect to find in the log
@@ -520,12 +522,21 @@ static int do_one_pass(journal_t *journal,if (descr_csum_size >0 && !jbd2_descriptor_block_csum_verify (journal, bh->b_data)) { -printk (KERN_ERR "JBD2: Invalid checksum " -"recovering block %lu in log\n" , - next_log_block); - err = -EFSBADCRC ; -brelse (bh); - goto failed; +/* +* PASS_SCAN can see stale blocks due to lazy +* journal init. Don't error out on those yet. +*/ +if (pass !=PASS_SCAN ) { +pr_err ("JBD2: Invalid checksum recovering block %lu in log\n" , + next_log_block); + err = -EFSBADCRC ; +brelse (bh); + goto failed; + } + need_check_commit_time = true; +jbd_debug (1 , +"invalid descriptor block found in %lu\n" , + next_log_block); }/* If it is a valid descriptor block, replay it
@@ -535,6 +546,7 @@ static int do_one_pass(journal_t *journal,if (pass !=PASS_REPLAY ) {if (pass ==PASS_SCAN &&jbd2_has_feature_checksum (journal) && + !need_check_commit_time && !info->end_transaction) { if (calc_chksums(journal, bh, &next_log_block,
@@ -683,11 +695,41 @@ static int do_one_pass(journal_t *journal,* mentioned conditions. Hence assume * "Interrupted Commit".) */ + commit_time =be64_to_cpu ( + ((struct commit_header *)bh->b_data)->h_commit_sec); +/* +* If need_check_commit_time is set, it means we are in +* PASS_SCAN and csum verify failed before. If +* commit_time is increasing, it's the same journal, +* otherwise it is stale journal block, just end this +* recovery. +*/ +if (need_check_commit_time) { +if (commit_time >= last_trans_commit_time) { +pr_err ("JBD2: Invalid checksum found in transaction %u\n" , + next_commit_ID); + err = -EFSBADCRC ; +brelse (bh); + goto failed; + } + ignore_crc_mismatch: +/* +* It likely does not belong to same journal, +* just end this recovery with success. +*/ +jbd_debug (1 ,"JBD2: Invalid checksum ignored in transaction %u, likely stale data\n" , + next_commit_ID); + err =0 ; +brelse (bh); + goto done; + } -/* Found an expected commit block: if checksums -* are present verify them in PASS_SCAN; else not +/* +* Found an expected commit block: if checksums +* are present, verify them in PASS_SCAN; else not * much to do other than move on to the next sequence -* number. */ +* number. +*/ if (pass ==PASS_SCAN &&jbd2_has_feature_checksum (journal)) { struct commit_header *cbh =
@@ -719,6 +761,8 @@ static int do_one_pass(journal_t *journal, !jbd2_commit_block_csum_verify(journal, bh->b_data)) { chksum_error: +if (commit_time < last_trans_commit_time) + goto ignore_crc_mismatch; info->end_transaction = next_commit_ID;if (!jbd2_has_feature_async_commit (journal)) {
@@ -728,11 +772,24 @@ static int do_one_pass(journal_t *journal,break ; } } +if (pass ==PASS_SCAN ) + last_trans_commit_time = commit_time;brelse (bh); next_commit_ID++;continue ;case JBD2_REVOKE_BLOCK : +/* +* Check revoke block crc in pass_scan, if csum verify +* failed, check commit block time later. +*/ +if (pass ==PASS_SCAN && + !jbd2_descriptor_block_csum_verify (journal, + bh->b_data)) { +jbd_debug (1 ,"JBD2: invalid revoke block found in %lu\n" , + next_log_block); + need_check_commit_time = true; + }/* If we aren't in the REVOKE pass, then we can * just skip over this block. */ if (pass !=PASS_REVOKE ) {
@@ -800,9 +857,6 @@ static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, offset =sizeof (jbd2_journal_revoke_header_t); rcount =be32_to_cpu (header->r_count); -if (!jbd2_descriptor_block_csum_verify (journal, header)) -return -EFSBADCRC ; -if (jbd2_journal_has_csum_v2or3 (journal)) csum_size =sizeof (struct jbd2_journal_block_tail );if (rcount > journal->j_blocksize - csum_size)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 119/147 ext4: Detect already used quota file early
In reply to Sasha LevinFrom: Jan Kara <[email protected]>
[ Upstream commit e0770e91424f694b461141cbc99adf6b23006b60 ]
When we try to use file already used as a quota file again (for the same or different quota type), strange things can happen. At the very least lockdep annotations may be wrong but also inode flags may be wrongly set / reset. When the file is used for two quota types at once we can even corrupt the file and likely crash the kernel. Catch all these cases by checking whether passed file is already used as quota file and bail early in that case.
This fixes occasional generic/219 failure due to lockdep complaint.
--- fs/ext4/super.c | 5 +++++ 1 file changed, 5 insertions(+)
fs/ext4/super.c +5 -0
@@ -6042,6 +6042,11 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id,/* Quotafile not on the same filesystem? */ if (path->dentry->d_sb != sb)return -EXDEV ; + +/* Quota already enabled for this file? */ +if (IS_NOQUOTA (d_inode (path->dentry))) +return -EBUSY ; +/* Journaling quota? */ if (EXT4_SB (sb)->s_qf_names[type]) {/* Quotafile not in fs root? */
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 120/147 PCI: dwc: Add link up check in dw_child_pcie_ops.map_bus()
In reply to Sasha LevinFrom: Hou Zhiqiang <[email protected]>
[ Upstream commit 15b23906347c0af8778d1d4edeea415290342d35 ]
NXP Layerscape (ls1028a, ls2088a), dra7xxx and imx6 platforms are either programmed or statically configured to forward the error triggered by a link-down state (eg no connected endpoint device) on the system bus for PCI configuration transactions; these errors are reported as an SError at system level, which is fatal.
Enumerating a PCI tree when the PCIe link is down is not sensible either, so even if the link-up check is racy (link can go down after map_bus() is called) add a link-up check in map_bus() to prevent issuing configuration transactions when the link is down.
SError report:
SError Interrupt on CPU2, code 0xbf000002 -- SError
CPU: 2 PID: 1 Comm: swapper/0 Not tainted 5.9.0-rc5-next-20200914-00001-gf965d3ec86fa #67 Hardware name: LS1046A RDB Board (DT) pstate: 20000085 (nzCv daIf -PAN -UAO BTYPE=--) pc : pci_generic_config_read+0x3c/0xe0 lr : pci_generic_config_read+0x24/0xe0 sp : ffff80001003b7b0 x29: ffff80001003b7b0 x28: ffff80001003ba74 x27: ffff000971d96800 x26: ffff00096e77e0a8 x25: ffff80001003b874 x24: ffff80001003b924 x23: 0000000000000004 x22: 0000000000000000 x21: 0000000000000000 x20: ffff80001003b874 x19: 0000000000000004 x18: ffffffffffffffff x17: 00000000000000c0 x16: fffffe0025981840 x15: ffffb94c75b69948 x14: 62203a383634203a x13: 666e6f635f726568 x12: 202c31203d207265 x11: 626d756e3e2d7375 x10: 656877202c307830 x9 : 203d206e66766564 x8 : 0000000000000908 x7 : 0000000000000908 x6 : ffff800010900000 x5 : ffff00096e77e080 x4 : 0000000000000000 x3 : 0000000000000003 x2 : 84fa3440ff7e7000 x1 : 0000000000000000 x0 : ffff800010034000 Kernel panic - not syncing: Asynchronous SError Interrupt CPU: 2 PID: 1 Comm: swapper/0 Not tainted 5.9.0-rc5-next-20200914-00001-gf965d3ec86fa #67 Hardware name: LS1046A RDB Board (DT) Call trace: dump_backtrace+0x0/0x1c0 show_stack+0x18/0x28 dump_stack+0xd8/0x134 panic+0x180/0x398 add_taint+0x0/0xb0 arm64_serror_panic+0x78/0x88 do_serror+0x68/0x180 el1_error+0x84/0x100 pci_generic_config_read+0x3c/0xe0 dw_pcie_rd_other_conf+0x78/0x110 pci_bus_read_config_dword+0x88/0xe8 pci_bus_generic_read_dev_vendor_id+0x30/0x1b0 pci_bus_read_dev_vendor_id+0x4c/0x78 pci_scan_single_device+0x80/0x100
--- drivers/pci/controller/dwc/pcie-designware-host.c | 11 +++++++++++ 1 file changed, 11 insertions(+)
drivers/pci/controller/dwc/pcie-designware-host.c +11 -0
@@ -515,6 +515,17 @@ static int dw_pcie_access_other_conf(struct pcie_port *pp, struct pci_bus *bus,void __iomem *va_cfg_base;struct dw_pcie *pci =to_dw_pcie_from_pp (pp); +/* +* Checking whether the link is up here is a last line of defense +* against platforms that forward errors on the system bus as +* SError upon PCI configuration transactions issued when the link +* is down. This check is racy by definition and does not stop +* the system from triggering an SError if the link goes down +* after this check is performed. +*/ +if (!dw_pcie_link_up (pci)) +return NULL ; + busdev =PCIE_ATU_BUS (bus->number) |PCIE_ATU_DEV (PCI_SLOT (devfn)) |PCIE_ATU_FUNC (PCI_FUNC (devfn));
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 121/147 KVM: PPC: Book3S HV: Do not allocate HPT for a nested guest
In reply to Sasha LevinFrom: Fabiano Rosas <[email protected]>
[ Upstream commit 05e6295dc7de859c9d56334805485c4d20bebf25 ]
The current nested KVM code does not support HPT guests. This is informed/enforced in some ways:
- Hosts < P9 will not be able to enable the nested HV feature;
- The nested hypervisor MMU capabilities will not contain KVM_CAP_PPC_MMU_HASH_V3;
- QEMU reflects the MMU capabilities in the 'ibm,arch-vec-5-platform-support' device-tree property;
- The nested guest, at 'prom_parse_mmu_model' ignores the 'disable_radix' kernel command line option if HPT is not supported;
- The KVM_PPC_CONFIGURE_V3_MMU ioctl will fail if trying to use HPT.
There is, however, still a way to start a HPT guest by using max-compat-cpu=power8 at the QEMU machine options. This leads to the guest being set to use hash after QEMU calls the KVM_PPC_ALLOCATE_HTAB ioctl.
With the guest set to hash, the nested hypervisor goes through the entry path that has no knowledge of nesting (kvmppc_run_vcpu) and crashes when it tries to execute an hypervisor-privileged (mtspr HDEC) instruction at __kvmppc_vcore_entry:
root@L1:~ $ qemu-system-ppc64 -machine pseries,max-cpu-compat=power8 ...
<snip>
[ 538.543303] CPU: 83 PID: 25185 Comm: CPU 0/KVM Not tainted 5.9.0-rc4 #1
[ 538.543355] NIP: c00800000753f388 LR: c00800000753f368 CTR: c0000000001e5ec0
[ 538.543417] REGS: c0000013e91e33b0 TRAP: 0700 Not tainted (5.9.0-rc4)
[ 538.543470] MSR: 8000000002843033 <SF,VEC,VSX,FP,ME,IR,DR,RI,LE> CR: 22422882 XER: 20040000
[ 538.543546] CFAR: c00800000753f4b0 IRQMASK: 3
GPR00: c0080000075397a0 c0000013e91e3640 c00800000755e600 0000000080000000
GPR04: 0000000000000000 c0000013eab19800 c000001394de0000 00000043a054db72
GPR08: 00000000003b1652 0000000000000000 0000000000000000 c0080000075502e0
GPR12: c0000000001e5ec0 c0000007ffa74200 c0000013eab19800 0000000000000008
GPR16: 0000000000000000 c00000139676c6c0 c000000001d23948 c0000013e91e38b8
GPR20: 0000000000000053 0000000000000000 0000000000000001 0000000000000000
GPR24: 0000000000000001 0000000000000001 0000000000000000 0000000000000001
GPR28: 0000000000000001 0000000000000053 c0000013eab19800 0000000000000001
[ 538.544067] NIP [c00800000753f388] __kvmppc_vcore_entry+0x90/0x104 [kvm_hv]
[ 538.544121] LR [c00800000753f368] __kvmppc_vcore_entry+0x70/0x104 [kvm_hv]
[ 538.544173] Call Trace:
[ 538.544196] [c0000013e91e3640] [c0000013e91e3680] 0xc0000013e91e3680 (unreliable)
[ 538.544260] [c0000013e91e3820] [c0080000075397a0] kvmppc_run_core+0xbc8/0x19d0 [kvm_hv]
[ 538.544325] [c0000013e91e39e0] [c00800000753d99c] kvmppc_vcpu_run_hv+0x404/0xc00 [kvm_hv]
[ 538.544394] [c0000013e91e3ad0] [c0080000072da4fc] kvmppc_vcpu_run+0x34/0x48 [kvm]
[ 538.544472] [c0000013e91e3af0] [c0080000072d61b8] kvm_arch_vcpu_ioctl_run+0x310/0x420 [kvm]
[ 538.544539] [c0000013e91e3b80] [c0080000072c7450] kvm_vcpu_ioctl+0x298/0x778 [kvm]
[ 538.544605] [c0000013e91e3ce0] [c0000000004b8c2c] sys_ioctl+0x1dc/0xc90
[ 538.544662] [c0000013e91e3dc0] [c00000000002f9a4] system_call_exception+0xe4/0x1c0
[ 538.544726] [c0000013e91e3e20] [c00000000000d140] system_call_common+0xf0/0x27c
[ 538.544787] Instruction dump:
[ 538.544821] f86d1098 60000000 60000000 48000099 e8ad0fe8 e8c500a0 e9264140 75290002
[ 538.544886] 7d1602a6 7cec42a6 40820008 7d0807b4 <7d164ba6> 7d083a14 f90d10a0 480104fd
[ 538.544953] ---[ end trace 74423e2b948c2e0c ]---This patch makes the KVM_PPC_ALLOCATE_HTAB ioctl fail when running in the nested hypervisor, causing QEMU to abort.
--- arch/powerpc/kvm/book3s_hv.c | 6 ++++++ 1 file changed, 6 insertions(+)
arch/powerpc/kvm/book3s_hv.c +6 -0
@@ -5257,6 +5257,12 @@ static long kvm_arch_vm_ioctl_hv(struct file *filp,case KVM_PPC_ALLOCATE_HTAB : {u32 htab_order; +/* If we're a nested hypervisor, we currently only support radix */ +if (kvmhv_on_pseries ()) { + r = -EOPNOTSUPP ; +break ; + } + r = -EFAULT ;if (get_user (htab_order, (u32 __user *)argp))break ;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 122/147 scsi: core: Clean up allocation and freeing of sgtables
In reply to Sasha LevinFrom: Christoph Hellwig <[email protected]>
[ Upstream commit 7007e9dd56767a95de0947b3f7599bcc2f21687f ]
Rename scsi_init_io() to scsi_alloc_sgtables(), and ensure callers call scsi_free_sgtables() to cleanup failures close to scsi_init_io() instead of leaking it down the generic I/O submission path.
--- drivers/scsi/scsi_lib.c | 22 ++++++++-------------- drivers/scsi/sd.c | 27 +++++++++++++++------------ drivers/scsi/sr.c | 16 ++++++---------- include/scsi/scsi_cmnd.h | 3 ++- 4 files changed, 31 insertions(+), 37 deletions(-)
drivers/scsi/scsi_lib.c +8 -14
@@ -530,7 +530,7 @@ static void scsi_uninit_cmd(struct scsi_cmnd *cmd) } } -static void scsi_free_sgtables (struct scsi_cmnd *cmd) +void scsi_free_sgtables (struct scsi_cmnd *cmd) {if (cmd->sdb.table.nents) sg_free_table_chained(&cmd->sdb.table,
@@ -539,6 +539,7 @@ static void scsi_free_sgtables(struct scsi_cmnd *cmd)sg_free_table_chained (&cmd->prot_sdb->table,SCSI_INLINE_PROT_SG_CNT ); } +EXPORT_SYMBOL_GPL (scsi_free_sgtables);static void scsi_mq_uninit_cmd (struct scsi_cmnd *cmd) {
@@ -966,7 +967,7 @@ static inline bool scsi_cmd_needs_dma_drain(struct scsi_device *sdev, }/** -* scsi_init_io - SCSI I/O initialization function. +* scsi_alloc_sgtables - allocate S/G tables for a command * @cmd: command descriptor we wish to initialize * * Returns:
@@ -974,7 +975,7 @@ static inline bool scsi_cmd_needs_dma_drain(struct scsi_device *sdev,* * BLK_STS_RESOURCE - if the failure is retryable * * BLK_STS_IOERR - if the failure is fatal */ -blk_status_t scsi_init_io (struct scsi_cmnd *cmd) +blk_status_t scsi_alloc_sgtables (struct scsi_cmnd *cmd) {struct scsi_device *sdev = cmd->device;struct request *rq = cmd->request;
@@ -1066,7 +1067,7 @@ blk_status_t scsi_init_io(struct scsi_cmnd *cmd)scsi_free_sgtables (cmd);return ret; } -EXPORT_SYMBOL (scsi_init_io); +EXPORT_SYMBOL (scsi_alloc_sgtables);/** * scsi_initialize_rq - initialize struct scsi_cmnd partially
@@ -1154,7 +1155,7 @@ static blk_status_t scsi_setup_scsi_cmnd(struct scsi_device *sdev,* submit a request without an attached bio. */ if (req->bio) { -blk_status_t ret =scsi_init_io (cmd); +blk_status_t ret =scsi_alloc_sgtables (cmd);if (unlikely (ret !=BLK_STS_OK ))return ret; }else {
@@ -1194,7 +1195,6 @@ static blk_status_t scsi_setup_cmnd(struct scsi_device *sdev, struct request *req) {struct scsi_cmnd *cmd =blk_mq_rq_to_pdu (req); -blk_status_t ret;if (!blk_rq_bytes (req)) cmd->sc_data_direction =DMA_NONE ;
@@ -1204,14 +1204,8 @@ static blk_status_t scsi_setup_cmnd(struct scsi_device *sdev, cmd->sc_data_direction =DMA_FROM_DEVICE ;if (blk_rq_is_scsi (req)) - ret =scsi_setup_scsi_cmnd (sdev, req); -else - ret =scsi_setup_fs_cmnd (sdev, req); - -if (ret !=BLK_STS_OK ) -scsi_free_sgtables (cmd); - -return ret; +return scsi_setup_scsi_cmnd (sdev, req); +return scsi_setup_fs_cmnd (sdev, req); } static blk_status_t
drivers/scsi/sd.c +15 -12
@@ -866,7 +866,7 @@ static blk_status_t sd_setup_unmap_cmnd(struct scsi_cmnd *cmd) cmd->transfersize = data_len; rq->timeout =SD_TIMEOUT ; -return scsi_init_io (cmd); +return scsi_alloc_sgtables (cmd); } static blk_status_t sd_setup_write_same16_cmnd(struct scsi_cmnd *cmd,
@@ -897,7 +897,7 @@ static blk_status_t sd_setup_write_same16_cmnd(struct scsi_cmnd *cmd, cmd->transfersize = data_len; rq->timeout = unmap ?SD_TIMEOUT :SD_WRITE_SAME_TIMEOUT ; -return scsi_init_io (cmd); +return scsi_alloc_sgtables (cmd); } static blk_status_t sd_setup_write_same10_cmnd(struct scsi_cmnd *cmd,
@@ -928,7 +928,7 @@ static blk_status_t sd_setup_write_same10_cmnd(struct scsi_cmnd *cmd, cmd->transfersize = data_len; rq->timeout = unmap ?SD_TIMEOUT :SD_WRITE_SAME_TIMEOUT ; -return scsi_init_io (cmd); +return scsi_alloc_sgtables (cmd); }static blk_status_t sd_setup_write_zeroes_cmnd (struct scsi_cmnd *cmd)
@@ -1069,7 +1069,7 @@ static blk_status_t sd_setup_write_same_cmnd(struct scsi_cmnd *cmd)* knows how much to actually write. */ rq->__data_len = sdp->sector_size; - ret =scsi_init_io (cmd); + ret =scsi_alloc_sgtables (cmd); rq->__data_len =blk_rq_bytes (rq);return ret;
@@ -1187,23 +1187,24 @@ static blk_status_t sd_setup_read_write_cmnd(struct scsi_cmnd *cmd)unsigned int dif;bool dix; - ret =scsi_init_io (cmd); + ret =scsi_alloc_sgtables (cmd);if (ret !=BLK_STS_OK )return ret; + ret =BLK_STS_IOERR ;if (!scsi_device_online (sdp) || sdp->changed) {scmd_printk (KERN_ERR , cmd,"device offline or changed\n" ); -return BLK_STS_IOERR ; + goto fail; }if (blk_rq_pos (rq) +blk_rq_sectors (rq) >get_capacity (rq->rq_disk)) {scmd_printk (KERN_ERR , cmd,"access beyond end of device\n" ); -return BLK_STS_IOERR ; + goto fail; }if ((blk_rq_pos (rq) & mask) || (blk_rq_sectors (rq) & mask)) {scmd_printk (KERN_ERR , cmd,"request not aligned to the logical block size\n" ); -return BLK_STS_IOERR ; + goto fail; }/*
@@ -1225,7 +1226,7 @@ static blk_status_t sd_setup_read_write_cmnd(struct scsi_cmnd *cmd)if (req_op (rq) ==REQ_OP_ZONE_APPEND ) { ret =sd_zbc_prepare_zone_append (cmd, &lba, nr_blocks);if (ret) -return ret; + goto fail; } fua = rq->cmd_flags &REQ_FUA ?0x8 :0 ;
@@ -1253,7 +1254,7 @@ static blk_status_t sd_setup_read_write_cmnd(struct scsi_cmnd *cmd) }if (unlikely (ret !=BLK_STS_OK )) -return ret; + goto fail;/* * We shouldn't disconnect in the middle of a sector, so with a dumb
@@ -1277,10 +1278,12 @@ static blk_status_t sd_setup_read_write_cmnd(struct scsi_cmnd *cmd) blk_rq_sectors(rq)));/* -* This indicates that the command is ready from our end to be -* queued. +* This indicates that the command is ready from our end to be queued. */ return BLK_STS_OK ; +fail: +scsi_free_sgtables (cmd); +return ret; }static blk_status_t sd_init_command (struct scsi_cmnd *cmd)
drivers/scsi/sr.c +6 -10
@@ -392,15 +392,11 @@ static blk_status_t sr_init_command(struct scsi_cmnd *SCpnt)struct request *rq = SCpnt->request;blk_status_t ret; - ret =scsi_init_io (SCpnt); + ret =scsi_alloc_sgtables (SCpnt);if (ret !=BLK_STS_OK ) - goto out; +return ret; cd =scsi_cd (rq->rq_disk); -/* from here on until we're complete, any goto out -* is used for a killable error condition */ - ret =BLK_STS_IOERR ; -SCSI_LOG_HLQUEUE (1 ,scmd_printk (KERN_INFO , SCpnt,"Doing sr request, block = %d\n" , block));
@@ -509,12 +505,12 @@ static blk_status_t sr_init_command(struct scsi_cmnd *SCpnt) SCpnt->allowed =MAX_RETRIES ;/* -* This indicates that the command is ready from our end to be -* queued. +* This indicates that the command is ready from our end to be queued. */ - ret =BLK_STS_OK ; +return BLK_STS_OK ; out: -return ret; +scsi_free_sgtables (SCpnt); +return BLK_STS_IOERR ; }static int sr_block_open (struct block_device *bdev,fmode_t mode)
include/scsi/scsi_cmnd.h +2 -1
@@ -165,7 +165,8 @@ extern void *scsi_kmap_atomic_sg(struct scatterlist *sg, int sg_count, size_t *offset, size_t *len);extern void scsi_kunmap_atomic_sg (void *virt); -extern blk_status_t scsi_init_io (struct scsi_cmnd *cmd); +blk_status_t scsi_alloc_sgtables (struct scsi_cmnd *cmd); +void scsi_free_sgtables (struct scsi_cmnd *cmd);#ifdef CONFIG_SCSI_DMA extern int scsi_dma_map (struct scsi_cmnd *cmd);
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 123/147 gfs2: call truncate_inode_pages_final for address space glocks
In reply to Sasha LevinFrom: Bob Peterson <[email protected]>
[ Upstream commit ee1e2c773e4f4ce2213f9d77cc703b669ca6fa3f ]
Before this patch, we were not calling truncate_inode_pages_final for the address space for glocks, which left the possibility of a leak. We now take care of the problem instead of complaining, and we do it during glock tear-down..
--- fs/gfs2/glock.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
fs/gfs2/glock.c +6 -1
@@ -270,7 +270,12 @@ static void __gfs2_glock_put(struct gfs2_glock *gl)gfs2_glock_remove_from_lru (gl);spin_unlock (&gl->gl_lockref.lock);GLOCK_BUG_ON (gl, !list_empty (&gl->gl_holders)); -GLOCK_BUG_ON (gl, mapping && mapping->nrpages && !gfs2_withdrawn (sdp)); +if (mapping) { +truncate_inode_pages_final (mapping); +if (!gfs2_withdrawn (sdp)) +GLOCK_BUG_ON (gl, mapping->nrpages || + mapping->nrexceptional); + }trace_gfs2_glock_put (gl); sdp->sd_lockstruct.ls_ops->lm_put_lock (gl); }
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 124/147 gfs2: Fix NULL pointer dereference in gfs2_rgrp_dump
In reply to Sasha LevinFrom: Andrew Price <[email protected]>
[ Upstream commit 0e539ca1bbbe85a86549c97a30a765ada4a09df9 ]
When an rindex entry is found to be corrupt, compute_bitstructs() calls gfs2_consist_rgrpd() which calls gfs2_rgrp_dump() like this:
gfs2_rgrp_dump(NULL, rgd->rd_gl, fs_id_buf);
gfs2_rgrp_dump then dereferences the gl without checking it and we get
BUG: KASAN: null-ptr-deref in gfs2_rgrp_dump+0x28/0x280
because there's no rgrp glock involved while reading the rindex on mount.
Fix this by changing gfs2_rgrp_dump to take an rgrp argument.
--- fs/gfs2/glops.c | 11 ++++++++++- fs/gfs2/rgrp.c | 9 +++------ fs/gfs2/rgrp.h | 2 +- fs/gfs2/util.c | 2 +- 4 files changed, 15 insertions(+), 9 deletions(-)
fs/gfs2/glops.c +10 -1
@@ -227,6 +227,15 @@ static void rgrp_go_inval(struct gfs2_glock *gl, int flags) rgd->rd_flags &= ~GFS2_RDF_UPTODATE ; } +static void gfs2_rgrp_go_dump (struct seq_file *seq,struct gfs2_glock *gl, +const char *fs_id_buf) +{ +struct gfs2_rgrpd *rgd =gfs2_glock2rgrp (gl); + +if (rgd) +gfs2_rgrp_dump (seq, rgd, fs_id_buf); +} +static struct gfs2_inode *gfs2_glock2inode (struct gfs2_glock *gl) {struct gfs2_inode *ip;
@@ -712,7 +721,7 @@ const struct gfs2_glock_operations gfs2_rgrp_glops = { .go_sync = rgrp_go_sync, .go_inval = rgrp_go_inval, .go_lock = gfs2_rgrp_go_lock, - .go_dump = gfs2_rgrp_dump, + .go_dump = gfs2_rgrp_go_dump, .go_type = LM_TYPE_RGRP, .go_flags = GLOF_LVB, };
fs/gfs2/rgrp.c +3 -6
@@ -2209,20 +2209,17 @@ static void rgblk_free(struct gfs2_sbd *sdp, struct gfs2_rgrpd *rgd,/** * gfs2_rgrp_dump - print out an rgrp * @seq: The iterator -* @gl: The glock in question +* @rgd: The rgrp in question * @fs_id_buf: pointer to file system id (if requested) * */ -void gfs2_rgrp_dump (struct seq_file *seq,struct gfs2_glock *gl, +void gfs2_rgrp_dump (struct seq_file *seq,struct gfs2_rgrpd *rgd,const char *fs_id_buf) { -struct gfs2_rgrpd *rgd = gl->gl_object;struct gfs2_blkreserv *trs;const struct rb_node *n; -if (rgd ==NULL ) -return ; gfs2_print_dbg(seq, "%s R: n:%llu f:%02x b:%u/%u i:%u r:%u e:%u\n", fs_id_buf, (unsigned long long)rgd->rd_addr, rgd->rd_flags,
@@ -2253,7 +2250,7 @@ static void gfs2_rgrp_error(struct gfs2_rgrpd *rgd) (unsigned long long)rgd->rd_addr);fs_warn (sdp,"umount on all nodes and run fsck.gfs2 to fix the error\n" );sprintf (fs_id_buf,"fsid=%s: " , sdp->sd_fsname); -gfs2_rgrp_dump (NULL , rgd->rd_gl, fs_id_buf); +gfs2_rgrp_dump (NULL , rgd, fs_id_buf); rgd->rd_flags |=GFS2_RDF_ERROR ; }
fs/gfs2/rgrp.h +1 -1
@@ -67,7 +67,7 @@ extern void gfs2_rlist_add(struct gfs2_inode *ip, struct gfs2_rgrp_list *rlist,extern void gfs2_rlist_alloc (struct gfs2_rgrp_list *rlist);extern void gfs2_rlist_free (struct gfs2_rgrp_list *rlist);extern u64 gfs2_ri_total (struct gfs2_sbd *sdp); -extern void gfs2_rgrp_dump (struct seq_file *seq,struct gfs2_glock *gl, +extern void gfs2_rgrp_dump (struct seq_file *seq,struct gfs2_rgrpd *rgd,const char *fs_id_buf); extern int gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset, struct buffer_head *bh,
fs/gfs2/util.c +1 -1
@@ -419,7 +419,7 @@ void gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd,char fs_id_buf[sizeof (sdp->sd_fsname) +7 ];sprintf (fs_id_buf,"fsid=%s: " , sdp->sd_fsname); -gfs2_rgrp_dump (NULL , rgd->rd_gl, fs_id_buf); +gfs2_rgrp_dump (NULL , rgd, fs_id_buf);gfs2_lm (sdp,"fatal: filesystem consistency error\n" " RG = %llu\n"
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 125/147 gfs2: use-after-free in sysfs deregistration
In reply to Sasha LevinFrom: Jamie Iles <[email protected]>
[ Upstream commit c2a04b02c060c4858762edce4674d5cba3e5a96f ]
syzkaller found the following splat with CONFIG_DEBUG_KOBJECT_RELEASE=y:
Read of size 1 at addr ffff000028e896b8 by task kworker/1:2/228 CPU: 1 PID: 228 Comm: kworker/1:2 Tainted: G S 5.9.0-rc8+ #101 Hardware name: linux,dummy-virt (DT) Workqueue: events kobject_delayed_cleanup Call trace: dump_backtrace+0x0/0x4d8 show_stack+0x34/0x48 dump_stack+0x174/0x1f8 print_address_description.constprop.0+0x5c/0x550 kasan_report+0x13c/0x1c0 __asan_report_load1_noabort+0x34/0x60 memcmp+0xd0/0xd8 gfs2_uevent+0xc4/0x188 kobject_uevent_env+0x54c/0x1240 kobject_uevent+0x2c/0x40 __kobject_del+0x190/0x1d8 kobject_delayed_cleanup+0x2bc/0x3b8 process_one_work+0x96c/0x18c0 worker_thread+0x3f0/0xc30 kthread+0x390/0x498 ret_from_fork+0x10/0x18 Allocated by task 1110: kasan_save_stack+0x28/0x58 __kasan_kmalloc.isra.0+0xc8/0xe8 kasan_kmalloc+0x10/0x20 kmem_cache_alloc_trace+0x1d8/0x2f0 alloc_super+0x64/0x8c0 sget_fc+0x110/0x620 get_tree_bdev+0x190/0x648 gfs2_get_tree+0x50/0x228 vfs_get_tree+0x84/0x2e8 path_mount+0x1134/0x1da8 do_mount+0x124/0x138 __arm64_sys_mount+0x164/0x238 el0_svc_common.constprop.0+0x15c/0x598 do_el0_svc+0x60/0x150 el0_svc+0x34/0xb0 el0_sync_handler+0xc8/0x5b4 el0_sync+0x15c/0x180 Freed by task 228: kasan_save_stack+0x28/0x58 kasan_set_track+0x28/0x40 kasan_set_free_info+0x24/0x48 __kasan_slab_free+0x118/0x190 kasan_slab_free+0x14/0x20 slab_free_freelist_hook+0x6c/0x210 kfree+0x13c/0x460
Use the same pattern as f2fs + ext4 where the kobject destruction must complete before allowing the FS itself to be freed. This means that we need an explicit free_sbd in the callers.
--- fs/gfs2/incore.h | 1 + fs/gfs2/ops_fstype.c | 22 +++++----------------- fs/gfs2/super.c | 1 + fs/gfs2/sys.c | 5 ++++- 4 files changed, 11 insertions(+), 18 deletions(-)
fs/gfs2/incore.h +1 -0
@@ -705,6 +705,7 @@ struct gfs2_sbd {struct super_block *sd_vfs;struct gfs2_pcpu_lkstats __percpu *sd_lkstats;struct kobject sd_kobj; +struct completion sd_kobj_unregister;unsigned long sd_flags;/* SDF_... */ struct gfs2_sb_host sd_sb;
fs/gfs2/ops_fstype.c +5 -17
@@ -1062,26 +1062,14 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc) } error =init_names (sdp, silent); -if (error) { -/* In this case, we haven't initialized sysfs, so we have to -manually free the sdp. */ -free_sbd (sdp); - sb->s_fs_info =NULL ; -return error; - } +if (error) + goto fail_free;snprintf (sdp->sd_fsname,sizeof (sdp->sd_fsname),"%s" , sdp->sd_table_name); error =gfs2_sys_fs_add (sdp); -/* -* If we hit an error here, gfs2_sys_fs_add will have called function -* kobject_put which causes the sysfs usage count to go to zero, which -* causes sysfs to call function gfs2_sbd_release, which frees sdp. -* Subsequent error paths here will call gfs2_sys_fs_del, which also -* kobject_put to free sdp. -*/ if (error) -return error; + goto fail_free;gfs2_create_debugfs_file (sdp);
@@ -1179,9 +1167,9 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc)gfs2_lm_unmount (sdp); fail_debug:gfs2_delete_debugfs_file (sdp); -/* gfs2_sys_fs_del must be the last thing we do, since it causes -* sysfs to call function gfs2_sbd_release, which frees sdp. */ gfs2_sys_fs_del (sdp); +fail_free: +free_sbd (sdp); sb->s_fs_info =NULL ;return error; }
fs/gfs2/super.c +1 -0
@@ -736,6 +736,7 @@ static void gfs2_put_super(struct super_block *sb)/* At this point, we're through participating in the lockspace */ gfs2_sys_fs_del (sdp); +free_sbd (sdp); }/**
fs/gfs2/sys.c +4 -1
@@ -303,7 +303,7 @@ static void gfs2_sbd_release(struct kobject *kobj) { struct gfs2_sbd *sdp = container_of(kobj, struct gfs2_sbd, sd_kobj); -free_sbd (sdp); +complete (&sdp->sd_kobj_unregister); }static struct kobj_type gfs2_ktype = {
@@ -655,6 +655,7 @@ int gfs2_sys_fs_add(struct gfs2_sbd *sdp)sprintf (ro,"RDONLY=%d" ,sb_rdonly (sb));sprintf (spectator,"SPECTATOR=%d" , sdp->sd_args.ar_spectator ?1 :0 ); +init_completion (&sdp->sd_kobj_unregister); sdp->sd_kobj.kset = gfs2_kset; error =kobject_init_and_add (&sdp->sd_kobj, &gfs2_ktype,NULL ,"%s" , sdp->sd_table_name);
@@ -685,6 +686,7 @@ int gfs2_sys_fs_add(struct gfs2_sbd *sdp) fail_reg:fs_err (sdp,"error %d adding sysfs files\n" , error);kobject_put (&sdp->sd_kobj); +wait_for_completion (&sdp->sd_kobj_unregister); sb->s_fs_info =NULL ;return error; }
@@ -695,6 +697,7 @@ void gfs2_sys_fs_del(struct gfs2_sbd *sdp)sysfs_remove_group (&sdp->sd_kobj, &tune_group);sysfs_remove_group (&sdp->sd_kobj, &lock_module_group);kobject_put (&sdp->sd_kobj); +wait_for_completion (&sdp->sd_kobj_unregister); } static int gfs2_uevent(struct kset *kset, struct kobject *kobj,
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 126/147 gfs2: add validation checks for size of superblock
In reply to Sasha LevinFrom: Anant Thazhemadam <[email protected]>
[ Upstream commit 0ddc5154b24c96f20e94d653b0a814438de6032b ]
In gfs2_check_sb(), no validation checks are performed with regards to the size of the superblock.
syzkaller detected a slab-out-of-bounds bug that was primarily caused because the block size for a superblock was set to zero.
A valid size for a superblock is a power of 2 between 512 and PAGE_SIZE. Performing validation checks and ensuring that the size of the superblock is valid fixes this bug.
--- fs/gfs2/ops_fstype.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-)
fs/gfs2/ops_fstype.c +11 -7
@@ -169,15 +169,19 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent)return -EINVAL ; } -/* If format numbers match exactly, we're done. */ - -if (sb->sb_fs_format ==GFS2_FORMAT_FS && - sb->sb_multihost_format ==GFS2_FORMAT_MULTI ) -return 0 ; +if (sb->sb_fs_format !=GFS2_FORMAT_FS || + sb->sb_multihost_format !=GFS2_FORMAT_MULTI ) { +fs_warn (sdp,"Unknown on-disk format, unable to mount\n" ); +return -EINVAL ; + } -fs_warn (sdp,"Unknown on-disk format, unable to mount\n" ); +if (sb->sb_bsize <512 || sb->sb_bsize >PAGE_SIZE || + (sb->sb_bsize & (sb->sb_bsize -1 ))) { +pr_warn ("Invalid superblock size\n" ); +return -EINVAL ; + } -return -EINVAL ; +return 0 ; }static void end_bio_io_page (struct bio *bio)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 127/147 Handle STATUS_IO_TIMEOUT gracefully
In reply to Sasha LevinFrom: Rohith Surabattula <[email protected]>
[ Upstream commit 8e670f77c4a55013db6d23b962f9bf6673a5e7b6 ]
Currently STATUS_IO_TIMEOUT is not treated as retriable error. It is currently mapped to ETIMEDOUT and returned to userspace for most system calls. STATUS_IO_TIMEOUT is returned by server in case of unavailability or throttling errors.
This patch will map the STATUS_IO_TIMEOUT to EAGAIN, so that it can be retried. Also, added a check to drop the connection to not overload the server in case of ongoing unavailability.
--- fs/cifs/cifsglob.h | 2 ++ fs/cifs/connect.c | 15 ++++++++++++++- fs/cifs/smb2maperror.c | 2 +- fs/cifs/smb2ops.c | 15 +++++++++++++++ 4 files changed, 32 insertions(+), 2 deletions(-)
fs/cifs/cifsglob.h +2 -0
@@ -510,6 +510,8 @@ struct smb_version_operations { struct fiemap_extent_info *, u64, u64);/* version specific llseek implementation */ loff_t (*llseek)(struct file *,struct cifs_tcon *,loff_t ,int ); +/* Check for STATUS_IO_TIMEOUT */ +bool (*is_status_io_timeout)(char *buf); };struct smb_version_values {
fs/cifs/connect.c +14 -1
@@ -69,6 +69,9 @@ extern bool disable_legacy_dialects;#define TLINK_ERROR_EXPIRE (1 * HZ)#define TLINK_IDLE_EXPIRE (600 * HZ) +/* Drop the connection to not overload the server */ +#define NUM_STATUS_IO_TIMEOUT 5 +enum {/* Mount options that take no arguments */ Opt_user_xattr, Opt_nouser_xattr,
@@ -1117,7 +1120,7 @@ cifs_demultiplex_thread(void *p)struct task_struct *task_to_wake =NULL ;struct mid_q_entry *mids[MAX_COMPOUND ];char *bufs[MAX_COMPOUND ]; -unsigned int noreclaim_flag; +unsigned int noreclaim_flag, num_io_timeout =0 ; noreclaim_flag =memalloc_noreclaim_save ();cifs_dbg (FYI ,"Demultiplex PID: %d\n" ,task_pid_nr (current));
@@ -1213,6 +1216,16 @@ cifs_demultiplex_thread(void *p)continue ; } +if (server->ops->is_status_io_timeout && + server->ops->is_status_io_timeout (buf)) { + num_io_timeout++; +if (num_io_timeout >NUM_STATUS_IO_TIMEOUT ) { +cifs_reconnect (server); + num_io_timeout =0 ; +continue ; + } + } + server->lstrp = jiffies;for (i =0 ; i < num_mids; i++) {
fs/cifs/smb2maperror.c +1 -1
@@ -488,7 +488,7 @@ static const struct status_to_posix_error smb2_error_map_table[] = { {STATUS_PIPE_CONNECTED, -EIO, "STATUS_PIPE_CONNECTED"}, {STATUS_PIPE_LISTENING, -EIO, "STATUS_PIPE_LISTENING"}, {STATUS_INVALID_READ_MODE, -EIO, "STATUS_INVALID_READ_MODE"}, - {STATUS_IO_TIMEOUT, -ETIMEDOUT, "STATUS_IO_TIMEOUT"}, + {STATUS_IO_TIMEOUT, -EAGAIN, "STATUS_IO_TIMEOUT"}, {STATUS_FILE_FORCED_CLOSED, -EIO, "STATUS_FILE_FORCED_CLOSED"}, {STATUS_PROFILING_NOT_STARTED, -EIO, "STATUS_PROFILING_NOT_STARTED"}, {STATUS_PROFILING_NOT_STOPPED, -EIO, "STATUS_PROFILING_NOT_STOPPED"},
fs/cifs/smb2ops.c +15 -0
@@ -2346,6 +2346,17 @@ smb2_is_session_expired(char *buf)return true; } +static bool +smb2_is_status_io_timeout (char *buf) +{ +struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; + +if (shdr->Status ==STATUS_IO_TIMEOUT ) +return true; +else +return false; +} +static int smb2_oplock_response (struct cifs_tcon *tcon,struct cifs_fid *fid,struct cifsInodeInfo *cinode)
@@ -4809,6 +4820,7 @@ struct smb_version_operations smb20_operations = { .make_node = smb2_make_node, .fiemap = smb3_fiemap, .llseek = smb3_llseek, + .is_status_io_timeout = smb2_is_status_io_timeout, };struct smb_version_operations smb21_operations = {
@@ -4909,6 +4921,7 @@ struct smb_version_operations smb21_operations = { .make_node = smb2_make_node, .fiemap = smb3_fiemap, .llseek = smb3_llseek, + .is_status_io_timeout = smb2_is_status_io_timeout, };struct smb_version_operations smb30_operations = {
@@ -5019,6 +5032,7 @@ struct smb_version_operations smb30_operations = { .make_node = smb2_make_node, .fiemap = smb3_fiemap, .llseek = smb3_llseek, + .is_status_io_timeout = smb2_is_status_io_timeout, };struct smb_version_operations smb311_operations = {
@@ -5130,6 +5144,7 @@ struct smb_version_operations smb311_operations = { .make_node = smb2_make_node, .fiemap = smb3_fiemap, .llseek = smb3_llseek, + .is_status_io_timeout = smb2_is_status_io_timeout, };struct smb_version_values smb20_values = {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 128/147 cifs: handle -EINTR in cifs_setattr
In reply to Sasha LevinFrom: Ronnie Sahlberg <[email protected]>
[ Upstream commit c6cc4c5a72505a0ecefc9b413f16bec512f38078 ]
RHBZ: 1848178
Some calls that set attributes, like utimensat(), are not supposed to return -EINTR and thus do not have handlers for this in glibc which causes us to leak -EINTR to the applications which are also unprepared to handle it.
For example tar will break if utimensat() return -EINTR and abort unpacking the archive. Other applications may break too.
To handle this we add checks, and retry, for -EINTR in cifs_setattr()
--- fs/cifs/inode.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
fs/cifs/inode.c +9 -4
@@ -2883,13 +2883,18 @@ cifs_setattr(struct dentry *direntry, struct iattr *attrs) {struct cifs_sb_info *cifs_sb =CIFS_SB (direntry->d_sb);struct cifs_tcon *pTcon =cifs_sb_master_tcon (cifs_sb); +int rc, retries =0 ; -if (pTcon->unix_ext) -return cifs_setattr_unix (direntry, attrs); - -return cifs_setattr_nounix (direntry, attrs); +do { +if (pTcon->unix_ext) + rc =cifs_setattr_unix (direntry, attrs); +else + rc =cifs_setattr_nounix (direntry, attrs); + retries++; + }while (is_retryable_error (rc) && retries <2 );/* BB: add cifs_setattr_legacy for really old servers */ +return rc; }#if 0
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 129/147 arm64: dts: renesas: ulcb: add full-pwr-cycle-in-suspend into eMMC nodes
In reply to Sasha LevinFrom: Yoshihiro Shimoda <[email protected]>
[ Upstream commit 992d7a8b88c83c05664b649fc54501ce58e19132 ]
Add full-pwr-cycle-in-suspend property to do a graceful shutdown of the eMMC device in system suspend.
--- arch/arm64/boot/dts/renesas/ulcb.dtsi | 1 + 1 file changed, 1 insertion(+)
arch/arm64/boot/dts/renesas/ulcb.dtsi +1 -0
@@ -469,6 +469,7 @@ &sdhi2 { mmc-hs200-1_8v; mmc-hs400-1_8v; non-removable; + full-pwr-cycle-in-suspend; status = "okay"; };
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 130/147 ARM: dts: omap4: Fix sgx clock rate for 4430
In reply to Sasha LevinFrom: Tony Lindgren <[email protected]>
[ Upstream commit 19d3e9a0bdd57b90175f30390edeb06851f5f9f3 ]
We currently have a different clock rate for droid4 compared to the stock v3.0.8 based Android Linux kernel:
# cat /sys/kernel/debug/clk/dpll_*_m7x2_ck/clk_rate 266666667 307200000 # cat /sys/kernel/debug/clk/l3_gfx_cm:clk:0000:0/clk_rate 307200000
Let's fix this by configuring sgx to use 153.6 MHz instead of 307.2 MHz. Looks like also at least duover needs this change to avoid hangs, so let's apply it for all 4430.
This helps a bit with thermal issues that seem to be related to memory corruption when using sgx. It seems that other driver related issues still remain though.
--- arch/arm/boot/dts/omap4.dtsi | 2 +- arch/arm/boot/dts/omap443x.dtsi | 10 ++++++++++ 2 files changed, 11 insertions(+), 1 deletion(-)
arch/arm/boot/dts/omap4.dtsi +1 -1
@@ -410,7 +410,7 @@ abb_iva: regulator-abb-iva { status = "disabled"; }; - target-module@56000000 { + sgx_module: target-module@56000000 { compatible = "ti,sysc-omap4", "ti,sysc"; reg = <0x5600fe00 0x4>, <0x5600fe10 0x4>;
arch/arm/boot/dts/omap443x.dtsi +10 -0
@@ -74,3 +74,13 @@ &cpu_thermal { }; /include/ "omap443x-clocks.dtsi" + +/* + * Use dpll_per for sgx at 153.6MHz like droid4 stock v3.0.8 Android kernel + */ +&sgx_module { + assigned-clocks = <&l3_gfx_clkctrl OMAP4_GPU_CLKCTRL 24>, + <&dpll_per_m7x2_ck>; + assigned-clock-rates = <0>, <153600000>; + assigned-clock-parents = <&dpll_per_m7x2_ck>; +};
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 131/147 soc: imx: gpcv2: Use dev_err_probe() to simplify error handling
In reply to Sasha LevinFrom: Anson Huang <[email protected]>
[ Upstream commit b663b798d04fb73f1ad4d54c46582d2fde7a76d6 ]
dev_err_probe() can reduce code size, uniform error handling and record the defer probe reason etc., use it to simplify the code.
--- drivers/soc/imx/gpcv2.c | 15 +++++---------- 1 file changed, 5 insertions(+), 10 deletions(-)
drivers/soc/imx/gpcv2.c +5 -10
@@ -487,22 +487,17 @@ static int imx_pgc_domain_probe(struct platform_device *pdev) domain->regulator =devm_regulator_get_optional (domain->dev,"power" );if (IS_ERR (domain->regulator)) { -if (PTR_ERR (domain->regulator) != -ENODEV ) { -if (PTR_ERR (domain->regulator) != -EPROBE_DEFER ) -dev_err (domain->dev,"Failed to get domain's regulator\n" ); -return PTR_ERR (domain->regulator); - } +if (PTR_ERR (domain->regulator) != -ENODEV ) +return dev_err_probe (domain->dev,PTR_ERR (domain->regulator), +"Failed to get domain's regulator\n" ); }else if (domain->voltage) {regulator_set_voltage (domain->regulator, domain->voltage, domain->voltage); } ret =imx_pgc_get_clocks (domain); -if (ret) { -if (ret != -EPROBE_DEFER ) -dev_err (domain->dev,"Failed to get domain's clocks\n" ); -return ret; - } +if (ret) +return dev_err_probe (domain->dev, ret,"Failed to get domain's clocks\n" ); ret =pm_genpd_init (&domain->genpd,NULL , true);if (ret) {
Signature · 7 lines
2.25.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Re: [PATCH AUTOSEL 5.9 131/147] soc: imx: gpcv2: Use dev_err_probe() to simplify error handling
Hi Sasha,
Quoted 5 lines · Sasha Levin
From: Anson Huang <[email protected]>
[ Upstream commit
b663b798d04fb73f1ad4d54c46582d2fde7a76d6]
Quoted 7 lines
dev_err_probe()can reduce code size, uniform error handling and record the defer probe reason etc., use it to simplify the code.Signed-off-by Anson Huang <[email protected]>Reviewed-by Guido Günther <[email protected]>Signed-off-by Shawn Guo <[email protected]>Signed-off-by Sasha Levin <[email protected]>
Does this qualify for stable since it is just a cleanup and not a bug fix?
Sasha Levin
maintainerRe: [PATCH AUTOSEL 5.9 131/147] soc: imx: gpcv2: Use dev_err_probe() to simplify error handling
Quoted 11 lines · Fabio Estevam
Hi Sasha,
Quoted 8 lines · Sasha Levin
From: Anson Huang <[email protected]>
[ Upstream commit
b663b798d04fb73f1ad4d54c46582d2fde7a76d6]
dev_err_probe()can reduce code size, uniform error handling and record the defer probe reason etc., use it to simplify the code.
Quoted 6 lines
Quoted 4 lines
Signed-off-by Anson Huang <[email protected]>Reviewed-by Guido Günther <[email protected]>Signed-off-by Shawn Guo <[email protected]>Signed-off-by Sasha Levin <[email protected]>Does this qualify for stable since it is just a cleanup and not a bug fix?
Nope, I'm dropping it, thanks!
Signature · 2 lines
Thanks, Sasha
Sasha Levin
maintainerPATCH 132/147 memory: emif: Remove bogus debugfs error handling
In reply to Sasha LevinFrom: Dan Carpenter <[email protected]>
[ Upstream commit fd22781648080cc400772b3c68aa6b059d2d5420 ]
Callers are generally not supposed to check the return values from debugfs functions. Debugfs functions never return NULL so this error handling will never trigger. (Historically debugfs functions used to return a mix of NULL and error pointers but it was eventually deemed too complicated for something which wasn't intended to be used in normal situations).
Delete all the error handling.
--- drivers/memory/emif.c | 33 +++++---------------------------- 1 file changed, 5 insertions(+), 28 deletions(-)
drivers/memory/emif.c +5 -28
@@ -163,35 +163,12 @@ static const struct file_operations emif_mr4_fops = { static int __init_or_module emif_debugfs_init(struct emif_data *emif) { -struct dentry *dentry; -int ret; - - dentry =debugfs_create_dir (dev_name (emif->dev),NULL ); -if (!dentry) { - ret = -ENOMEM ; - goto err0; - } - emif->debugfs_root = dentry; - - dentry =debugfs_create_file ("regcache_dump" ,S_IRUGO , - emif->debugfs_root, emif, &emif_regdump_fops); -if (!dentry) { - ret = -ENOMEM ; - goto err1; - } - - dentry =debugfs_create_file ("mr4" ,S_IRUGO , - emif->debugfs_root, emif, &emif_mr4_fops); -if (!dentry) { - ret = -ENOMEM ; - goto err1; - } - + emif->debugfs_root =debugfs_create_dir (dev_name (emif->dev),NULL ); +debugfs_create_file ("regcache_dump" ,S_IRUGO , emif->debugfs_root, emif, + &emif_regdump_fops); +debugfs_create_file ("mr4" ,S_IRUGO , emif->debugfs_root, emif, + &emif_mr4_fops);return 0 ; -err1: -debugfs_remove_recursive (emif->debugfs_root); -err0: -return ret; } static void __exit emif_debugfs_exit(struct emif_data *emif)
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 133/147 ARM: dts: s5pv210: Enable audio on Aries boards
In reply to Sasha LevinFrom: Jonathan Bakker <[email protected]>
[ Upstream commit cd972fe90008adf49de0790250c1275480ac5cdc ]
Both the Galaxy S and the Fascinate4G have a WM8994 codec, but they differ slightly in their jack detection and micbias configuration.
--- arch/arm/boot/dts/s5pv210-aries.dtsi | 10 +++ arch/arm/boot/dts/s5pv210-fascinate4g.dts | 98 +++++++++++++++++++++++ arch/arm/boot/dts/s5pv210-galaxys.dts | 85 ++++++++++++++++++++ 3 files changed, 193 insertions(+)
arch/arm/boot/dts/s5pv210-aries.dtsi +10 -0
@@ -47,6 +47,11 @@ mfc_right: region@51000000 { }; }; + bt_codec: bt_sco { + compatible = "linux,bt-sco"; + #sound-dai-cells = <0>; + }; + vibrator_pwr: regulator-fixed-0 { compatible = "regulator-fixed"; regulator-name = "vibrator-en";
@@ -624,6 +629,11 @@ touchscreen@4a { }; }; +&i2s0 { + dmas = <&pdma0 9>, <&pdma0 10>, <&pdma0 11>; + status = "okay"; +}; + &mfc { memory-region = <&mfc_left>, <&mfc_right>; };
arch/arm/boot/dts/s5pv210-fascinate4g.dts +98 -0
116 lines not on this page · raw message
arch/arm/boot/dts/s5pv210-galaxys.dts +85 -0
109 lines not on this page · raw message
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 134/147 ARM: dts: s5pv210: remove DMA controller bus node name to fix dtschema warnings
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit ea4e792f3c8931fffec4d700cf6197d84e9f35a6 ]
There is no need to keep DMA controller nodes under AMBA bus node. Remove the "amba" node to fix dtschema warnings like:
amba: $nodename:0: 'amba' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'
--- arch/arm/boot/dts/s5pv210.dtsi | 49 +++++++++++++++------------------- 1 file changed, 21 insertions(+), 28 deletions(-)
arch/arm/boot/dts/s5pv210.dtsi +21 -28
@@ -128,35 +128,28 @@ wakeup-interrupt-controller { }; }; - amba { - #address-cells = <1>; - #size-cells = <1>; - compatible = "simple-bus"; - ranges; - - pdma0: dma@e0900000 { - compatible = "arm,pl330", "arm,primecell"; - reg = <0xe0900000 0x1000>; - interrupt-parent = <&vic0>; - interrupts = <19>; - clocks = <&clocks CLK_PDMA0>; - clock-names = "apb_pclk"; - #dma-cells = <1>; - #dma-channels = <8>; - #dma-requests = <32>; - }; + pdma0: dma@e0900000 { + compatible = "arm,pl330", "arm,primecell"; + reg = <0xe0900000 0x1000>; + interrupt-parent = <&vic0>; + interrupts = <19>; + clocks = <&clocks CLK_PDMA0>; + clock-names = "apb_pclk"; + #dma-cells = <1>; + #dma-channels = <8>; + #dma-requests = <32>; + }; - pdma1: dma@e0a00000 { - compatible = "arm,pl330", "arm,primecell"; - reg = <0xe0a00000 0x1000>; - interrupt-parent = <&vic0>; - interrupts = <20>; - clocks = <&clocks CLK_PDMA1>; - clock-names = "apb_pclk"; - #dma-cells = <1>; - #dma-channels = <8>; - #dma-requests = <32>; - }; + pdma1: dma@e0a00000 { + compatible = "arm,pl330", "arm,primecell"; + reg = <0xe0a00000 0x1000>; + interrupt-parent = <&vic0>; + interrupts = <20>; + clocks = <&clocks CLK_PDMA1>; + clock-names = "apb_pclk"; + #dma-cells = <1>; + #dma-channels = <8>; + #dma-requests = <32>; }; adc: adc@e1700000 {
Signature · 7 lines
2.25.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Sasha Levin
maintainerPATCH 135/147 ARM: dts: s5pv210: move fixed clocks under root node
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit d38cae370e5f2094cbc38db3082b8e9509ae52ce ]
The fixed clocks are kept under dedicated 'external-clocks' node, thus a fake 'reg' was added. This is not correct with dtschema as fixed-clock binding does not have a 'reg' property. Moving fixed clocks out of 'soc' to root node fixes multiple dtbs_check warnings:
external-clocks: $nodename:0: 'external-clocks' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'
external-clocks: #size-cells:0:0: 0 is not one of [1, 2]
external-clocks: oscillator@0:reg:0: [0] is too short
external-clocks: oscillator@1:reg:0: [1] is too short
external-clocks: 'ranges' is a required property
oscillator@0: 'reg' does not match any of the regexes: 'pinctrl-[0-9]+'
--- arch/arm/boot/dts/s5pv210.dtsi | 36 +++++++++++++--------------------- 1 file changed, 14 insertions(+), 22 deletions(-)
arch/arm/boot/dts/s5pv210.dtsi +14 -22
@@ -52,34 +52,26 @@ cpu@0 { }; }; + xxti: oscillator-0 { + compatible = "fixed-clock"; + clock-frequency = <0>; + clock-output-names = "xxti"; + #clock-cells = <0>; + }; + + xusbxti: oscillator-1 { + compatible = "fixed-clock"; + clock-frequency = <0>; + clock-output-names = "xusbxti"; + #clock-cells = <0>; + }; + soc { compatible = "simple-bus"; #address-cells = <1>; #size-cells = <1>; ranges; - external-clocks { - compatible = "simple-bus"; - #address-cells = <1>; - #size-cells = <0>; - - xxti: oscillator@0 { - compatible = "fixed-clock"; - reg = <0>; - clock-frequency = <0>; - clock-output-names = "xxti"; - #clock-cells = <0>; - }; - - xusbxti: oscillator@1 { - compatible = "fixed-clock"; - reg = <1>; - clock-frequency = <0>; - clock-output-names = "xusbxti"; - #clock-cells = <0>; - }; - }; - onenand: onenand@b0600000 { compatible = "samsung,s5pv210-onenand"; reg = <0xb0600000 0x2000>,
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 136/147 ARM: dts: s5pv210: move PMU node out of clock controller
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit bb98fff84ad1ea321823759edaba573a16fa02bd ]
The Power Management Unit (PMU) is a separate device which has little common with clock controller. Moving it to one level up (from clock controller child to SoC) allows to remove fake simple-bus compatible and dtbs_check warnings like:
clock-controller@e0100000: $nodename:0:
'clock-controller@e0100000' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'
--- arch/arm/boot/dts/s5pv210.dtsi | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-)
arch/arm/boot/dts/s5pv210.dtsi +5 -8
@@ -92,19 +92,16 @@ chipid@e0000000 { }; clocks: clock-controller@e0100000 { - compatible = "samsung,s5pv210-clock", "simple-bus"; + compatible = "samsung,s5pv210-clock"; reg = <0xe0100000 0x10000>; clock-names = "xxti", "xusbxti"; clocks = <&xxti>, <&xusbxti>; #clock-cells = <1>; - #address-cells = <1>; - #size-cells = <1>; - ranges; + }; - pmu_syscon: syscon@e0108000 { - compatible = "samsung-s5pv210-pmu", "syscon"; - reg = <0xe0108000 0x8000>; - }; + pmu_syscon: syscon@e0108000 { + compatible = "samsung-s5pv210-pmu", "syscon"; + reg = <0xe0108000 0x8000>; }; pinctrl0: pinctrl@e0200000 {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 137/147 ARM: dts: s5pv210: remove dedicated 'audio-subsystem' node
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 6c17a2974abf68a58517f75741b15c4aba42b4b8 ]
The 'audio-subsystem' node is an artificial creation, not representing real hardware. The hardware is described by its nodes - AUDSS clock controller and I2S0.
Remove the 'audio-subsystem' node along with its undocumented compatible to fix dtbs_check warnings like:
audio-subsystem: $nodename:0: 'audio-subsystem' does not match '^([a-z][a-z0-9\\-]+-bus|bus|soc|axi|ahb|apb)(@[0-9a-f]+)?$'
--- arch/arm/boot/dts/s5pv210.dtsi | 65 +++++++++++++++------------------- 1 file changed, 29 insertions(+), 36 deletions(-)
arch/arm/boot/dts/s5pv210.dtsi +29 -36
@@ -223,43 +223,36 @@ i2c2: i2c@e1a00000 { status = "disabled"; }; - audio-subsystem { - compatible = "samsung,s5pv210-audss", "simple-bus"; - #address-cells = <1>; - #size-cells = <1>; - ranges; - - clk_audss: clock-controller@eee10000 { - compatible = "samsung,s5pv210-audss-clock"; - reg = <0xeee10000 0x1000>; - clock-names = "hclk", "xxti", - "fout_epll", - "sclk_audio0"; - clocks = <&clocks DOUT_HCLKP>, <&xxti>, - <&clocks FOUT_EPLL>, - <&clocks SCLK_AUDIO0>; - #clock-cells = <1>; - }; + clk_audss: clock-controller@eee10000 { + compatible = "samsung,s5pv210-audss-clock"; + reg = <0xeee10000 0x1000>; + clock-names = "hclk", "xxti", + "fout_epll", + "sclk_audio0"; + clocks = <&clocks DOUT_HCLKP>, <&xxti>, + <&clocks FOUT_EPLL>, + <&clocks SCLK_AUDIO0>; + #clock-cells = <1>; + }; - i2s0: i2s@eee30000 { - compatible = "samsung,s5pv210-i2s"; - reg = <0xeee30000 0x1000>; - interrupt-parent = <&vic2>; - interrupts = <16>; - dma-names = "rx", "tx", "tx-sec"; - dmas = <&pdma1 9>, <&pdma1 10>, <&pdma1 11>; - clock-names = "iis", - "i2s_opclk0", - "i2s_opclk1"; - clocks = <&clk_audss CLK_I2S>, - <&clk_audss CLK_I2S>, - <&clk_audss CLK_DOUT_AUD_BUS>; - samsung,idma-addr = <0xc0010000>; - pinctrl-names = "default"; - pinctrl-0 = <&i2s0_bus>; - #sound-dai-cells = <0>; - status = "disabled"; - }; + i2s0: i2s@eee30000 { + compatible = "samsung,s5pv210-i2s"; + reg = <0xeee30000 0x1000>; + interrupt-parent = <&vic2>; + interrupts = <16>; + dma-names = "rx", "tx", "tx-sec"; + dmas = <&pdma1 9>, <&pdma1 10>, <&pdma1 11>; + clock-names = "iis", + "i2s_opclk0", + "i2s_opclk1"; + clocks = <&clk_audss CLK_I2S>, + <&clk_audss CLK_I2S>, + <&clk_audss CLK_DOUT_AUD_BUS>; + samsung,idma-addr = <0xc0010000>; + pinctrl-names = "default"; + pinctrl-0 = <&i2s0_bus>; + #sound-dai-cells = <0>; + status = "disabled"; }; i2s1: i2s@e2100000 {
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 138/147 ARM: dts: s5pv210: add RTC 32 KHz clock in Aries family
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 086c4498b0cc87fdb09188f3da7056e898814948 ]
The S3C RTC requires 32768 Hz clock as input which is provided by PMIC. However there is no such clock provider but rather a regulator driver which registers the clock as a regulator. This is an old driver which will not be updated so add a workaround - a fixed-clock to fill missing clock phandle reference in S3C RTC.
This fixes dtbs_check warnings:
rtc@e2800000: clocks: [[2, 145]] is too short
rtc@e2800000: clock-names: ['rtc'] is too short
--- arch/arm/boot/dts/s5pv210-aries.dtsi | 12 ++++++++++++ 1 file changed, 12 insertions(+)
arch/arm/boot/dts/s5pv210-aries.dtsi +12 -0
@@ -47,6 +47,13 @@ mfc_right: region@51000000 { }; }; + pmic_ap_clk: clock-0 { + /* Workaround for missing clock on PMIC */ + compatible = "fixed-clock"; + #clock-cells = <0>; + clock-frequency = <32768>; + }; + bt_codec: bt_sco { compatible = "linux,bt-sco"; #sound-dai-cells = <0>;
@@ -825,6 +832,11 @@ &pwm { samsung,pwm-outputs = <1>; }; +&rtc { + clocks = <&clocks CLK_RTC>, <&pmic_ap_clk>; + clock-names = "rtc", "rtc_src"; +}; + &sdhci1 { #address-cells = <1>; #size-cells = <0>;
Signature · 7 lines
2.25.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
Sasha Levin
maintainerPATCH 139/147 ARM: dts: s5pv210: align SPI GPIO node name with dtschema in Aries
In reply to Sasha LevinFrom: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 1ed7f6d0bab2f1794f1eb4ed032e90575552fd21 ]
The device tree schema expects SPI controller to be named "spi", otherwise dtbs_check complain with a warning like:
spi-gpio-0: $nodename:0: 'spi-gpio-0' does not match '^spi(@.*|-[0-9a-f])*$'
--- arch/arm/boot/dts/s5pv210-aries.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/s5pv210-aries.dtsi +1 -1
@@ -545,7 +545,7 @@ poweroff: syscon-poweroff { value = <0x5200>; }; - spi_lcd: spi-gpio-0 { + spi_lcd: spi-2 { compatible = "spi-gpio"; #address-cells = <1>; #size-cells = <0>;
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 140/147 soc: qcom: rpmh-rsc: Sleep waiting for tcs slots to be free
In reply to Sasha LevinFrom: Stephen Boyd <[email protected]>
[ Upstream commit 2bc20f3c8487bd5bc4dd9ad2c06d2ba05fd4e838 ]
The busy loop in rpmh_rsc_send_data() is written with the assumption that the udelay will be preempted by the tcs_tx_done() irq handler when the TCS slots are all full. This doesn't hold true when the calling thread is an irqthread and the tcs_tx_done() irq is also an irqthread. That's because kernel irqthreads are SCHED_FIFO and thus need to voluntarily give up priority by calling into the scheduler so that other threads can run.
I see RCU stalls when I boot with irqthreads on the kernel commandline because the modem remoteproc driver is trying to send an rpmh async message from an irqthread that needs to give up the CPU for the rpmh irqthread to run and clear out tcs slots.
rcu: INFO: rcu_preempt self-detected stall on CPU rcu: 0-....: (1 GPs behind) idle=402/1/0x4000000000000002 softirq=2108/2109 fqs=4920 (t=21016 jiffies g=2933 q=590) Task dump for CPU 0: irq/11-smp2p R running task 0 148 2 0x00000028 Call trace: dump_backtrace+0x0/0x154 show_stack+0x20/0x2c sched_show_task+0xfc/0x108 dump_cpu_task+0x44/0x50 rcu_dump_cpu_stacks+0xa4/0xf8 rcu_sched_clock_irq+0x7dc/0xaa8 update_process_times+0x30/0x54 tick_sched_handle+0x50/0x64 tick_sched_timer+0x4c/0x8c __hrtimer_run_queues+0x21c/0x36c hrtimer_interrupt+0xf0/0x22c arch_timer_handler_phys+0x40/0x50 handle_percpu_devid_irq+0x114/0x25c __handle_domain_irq+0x84/0xc4 gic_handle_irq+0xd0/0x178 el1_irq+0xbc/0x180 save_return_addr+0x18/0x28 return_address+0x54/0x88 preempt_count_sub+0x40/0x88 _raw_spin_unlock_irqrestore+0x4c/0x6c ___ratelimit+0xd0/0x128 rpmh_rsc_send_data+0x24c/0x378 __rpmh_write+0x1b0/0x208 rpmh_write_async+0x90/0xbc rpmhpd_send_corner+0x60/0x8c rpmhpd_aggregate_corner+0x8c/0x124 rpmhpd_set_performance_state+0x8c/0xbc _genpd_set_performance_state+0xdc/0x1b8 dev_pm_genpd_set_performance_state+0xb8/0xf8 q6v5_pds_disable+0x34/0x60 [qcom_q6v5_mss] qcom_msa_handover+0x38/0x44 [qcom_q6v5_mss] q6v5_handover_interrupt+0x24/0x3c [qcom_q6v5] handle_nested_irq+0xd0/0x138 qcom_smp2p_intr+0x188/0x200 irq_thread_fn+0x2c/0x70 irq_thread+0xfc/0x14c kthread+0x11c/0x12c ret_from_fork+0x10/0x18
This busy loop naturally lends itself to using a wait queue so that each thread that tries to send a message will sleep waiting on the waitqueue and only be woken up when a free slot is available. This should make things more predictable too because the scheduler will be able to sleep tasks that are waiting on a free tcs instead of the busy loop we currently have today.
--- drivers/soc/qcom/rpmh-internal.h | 4 ++ drivers/soc/qcom/rpmh-rsc.c | 115 +++++++++++++++---------------- 2 files changed, 58 insertions(+), 61 deletions(-)
drivers/soc/qcom/rpmh-internal.h +4 -0
@@ -8,6 +8,7 @@#define __RPM_INTERNAL_H__#include <linux/bitmap.h> +#include <linux/wait.h> #include <soc/qcom/tcs.h> #define TCS_TYPE_NR 4
@@ -106,6 +107,8 @@ struct rpmh_ctrlr {* @lock: Synchronize state of the controller. If RPMH's cache * lock will also be held, the order is: drv->lock then * cache_lock. +* @tcs_wait: Wait queue used to wait for @tcs_in_use to free up a +* slot * @client: Handle to the DRV's client. */ struct rsc_drv {
@@ -118,6 +121,7 @@ struct rsc_drv {struct tcs_group tcs[TCS_TYPE_NR ];DECLARE_BITMAP (tcs_in_use,MAX_TCS_NR );spinlock_t lock; +wait_queue_head_t tcs_wait;struct rpmh_ctrlr client; };
drivers/soc/qcom/rpmh-rsc.c +54 -61
163 lines not on this page · raw message
Signature · 1 line
2.25.1
Sasha Levin
maintainerPATCH 141/147 soc: ti: k3: ringacc: add am65x sr2.0 support
In reply to Sasha LevinFrom: Grygorii Strashko <[email protected]>
[ Upstream commit 95e7be062aea6d2e09116cd4d28957d310c04781 ]
The AM65x SR2.0 Ringacc has fixed errata i2023 "RINGACC, UDMA: RINGACC and UDMA Ring State Interoperability Issue after Channel Teardown". This errata also fixed for J271E SoC.
Use SOC bus data for K3 SoC identification and enable i2023 errate w/a only for the AM65x SR1.0. This also makes obsolete "ti,dma-ring-reset-quirk" DT property.
--- drivers/soc/ti/k3-ringacc.c | 33 ++++++++++++++++++++++++++++++--- 1 file changed, 30 insertions(+), 3 deletions(-)
drivers/soc/ti/k3-ringacc.c +30 -3
@@ -10,6 +10,7 @@#include <linux/init.h> #include <linux/of.h> #include <linux/platform_device.h> +#include <linux/sys_soc.h> #include <linux/soc/ti/k3-ringacc.h> #include <linux/soc/ti/ti_sci_protocol.h> #include <linux/soc/ti/ti_sci_inta_msi.h>
@@ -208,6 +209,15 @@ struct k3_ringacc {const struct k3_ringacc_ops *ops; }; +/** +* struct k3_ringacc - Rings accelerator SoC data +* +* @dma_ring_reset_quirk: DMA reset w/a enable +*/ +struct k3_ringacc_soc_data { +unsigned dma_ring_reset_quirk:1 ; +}; +static long k3_ringacc_ring_get_fifo_pos (struct k3_ring *ring) { return K3_RINGACC_FIFO_WINDOW_SIZE_BYTES -
@@ -1051,9 +1061,6 @@ static int k3_ringacc_probe_dt(struct k3_ringacc *ringacc)return ret; } - ringacc->dma_ring_reset_quirk = -of_property_read_bool (node,"ti,dma-ring-reset-quirk" ); - ringacc->tisci =ti_sci_get_by_phandle (node,"ti,sci" );if (IS_ERR (ringacc->tisci)) { ret =PTR_ERR (ringacc->tisci);
@@ -1084,9 +1091,22 @@ static int k3_ringacc_probe_dt(struct k3_ringacc *ringacc) ringacc->rm_gp_range); } +static const struct k3_ringacc_soc_data k3_ringacc_soc_data_sr1 = { + .dma_ring_reset_quirk =1 , +}; + +static const struct soc_device_attribute k3_ringacc_socinfo[] = { + { .family ="AM65X" , + .revision ="SR1.0" , + .data = &k3_ringacc_soc_data_sr1 + }, + {/* sentinel */ } +}; +static int k3_ringacc_init (struct platform_device *pdev,struct k3_ringacc *ringacc) { +const struct soc_device_attribute *soc;void __iomem *base_fifo, *base_rt;struct device *dev = &pdev->dev;struct resource *res;
@@ -1103,6 +1123,13 @@ static int k3_ringacc_init(struct platform_device *pdev,if (ret)return ret; + soc =soc_device_match (k3_ringacc_socinfo); +if (soc && soc->data) { +const struct k3_ringacc_soc_data *soc_data = soc->data; + + ringacc->dma_ring_reset_quirk = soc_data->dma_ring_reset_quirk; + } + res =platform_get_resource_byname (pdev,IORESOURCE_MEM ,"rt" ); base_rt =devm_ioremap_resource (dev, res);if (IS_ERR (base_rt))
Signature · 7 lines
2.25.1 _______________________________________________ linux-arm-kernel mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-arm-kernel