RDMA/mlx4: Initialize ib_spec on the stack
Subsystems (18)
Sasha Levin
maintainerPATCH 1/35 RDMA/mlx4: Initialize ib_spec on the stack
From: Alaa Hleihel <[email protected]>
[ Upstream commit c08cfb2d8d78bfe81b37cc6ba84f0875bddd0d5c ]
Initialize ib_spec on the stack before using it, otherwise we will have garbage values that will break creating default rules with invalid parsing error.
--- drivers/infiniband/hw/mlx4/main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/infiniband/hw/mlx4/main.c +2 -1
@@ -1492,8 +1492,9 @@ static int __mlx4_ib_create_default_rules(int i;for (i =0 ; i <ARRAY_SIZE (pdefault_rules->rules_create_list); i++) { +union ib_flow_spec ib_spec = {};int ret; -union ib_flow_spec ib_spec; +switch (pdefault_rules->rules_create_list[i]) {case 0 :/* no rule */
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 2/35 RDMA/siw: Fix potential siw_mem refcnt leak in siw_fastreg_mr()
From: Jason Gunthorpe <[email protected]>
[ Upstream commit 6e051971b0e2eeb0ce7ec65d3cc8180450512d36 ]
siw_fastreg_mr() invokes siw_mem_id2obj(), which returns a local reference of the siw_mem object to "mem" with increased refcnt. When siw_fastreg_mr() returns, "mem" becomes invalid, so the refcount should be decreased to keep refcount balanced.
The issue happens in one error path of siw_fastreg_mr(). When "base_mr" equals to NULL but "mem" is not NULL, the function forgets to decrease the refcnt increased by siw_mem_id2obj() and causes a refcnt leak.
Reorganize the flow so that the goto unwind can be used as expected.
--- drivers/infiniband/sw/siw/siw_qp_tx.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-)
drivers/infiniband/sw/siw/siw_qp_tx.c +11 -4
@@ -920,20 +920,27 @@ static int siw_fastreg_mr(struct ib_pd *pd, struct siw_sqe *sqe) {struct ib_mr *base_mr = (struct ib_mr *)(uintptr_t )sqe->base_mr;struct siw_device *sdev =to_siw_dev (pd->device); -struct siw_mem *mem =siw_mem_id2obj (sdev, sqe->rkey >>8 ); +struct siw_mem *mem;int rv =0 ;siw_dbg_pd (pd,"STag 0x%08x\n" , sqe->rkey); -if (unlikely (!mem || !base_mr)) { +if (unlikely (!base_mr)) {pr_warn ("siw: fastreg: STag 0x%08x unknown\n" , sqe->rkey);return -EINVAL ; } +if (unlikely (base_mr->rkey >>8 != sqe->rkey >>8 )) {pr_warn ("siw: fastreg: STag 0x%08x: bad MR\n" , sqe->rkey); - rv = -EINVAL ; - goto out; +return -EINVAL ; } + + mem =siw_mem_id2obj (sdev, sqe->rkey >>8 ); +if (unlikely (!mem)) { +pr_warn ("siw: fastreg: STag 0x%08x unknown\n" , sqe->rkey); +return -EINVAL ; + } +if (unlikely (mem->pd != pd)) {pr_warn ("siw: fastreg: PD mismatch\n" ); rv = -EINVAL ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 3/35 nfs: Fix potential posix_acl refcnt leak in nfs3_set_acl
In reply to Sasha LevinFrom: Andreas Gruenbacher <[email protected]>
[ Upstream commit 7648f939cb919b9d15c21fff8cd9eba908d595dc ]
nfs3_set_acl keeps track of the acl it allocated locally to determine if an acl needs to be released at the end. This results in a memory leak when the function allocates an acl as well as a default acl. Fix by releasing acls that differ from the acl originally passed into nfs3_set_acl.
--- fs/nfs/nfs3acl.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-)
fs/nfs/nfs3acl.c +15 -7
@@ -253,37 +253,45 @@ int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl,int nfs3_set_acl (struct inode *inode,struct posix_acl *acl,int type) { -struct posix_acl *alloc =NULL , *dfacl =NULL ; +struct posix_acl *orig = acl, *dfacl =NULL , *alloc;int status;if (S_ISDIR (inode->i_mode)) {switch (type) {case ACL_TYPE_ACCESS : - alloc = dfacl =get_acl (inode,ACL_TYPE_DEFAULT ); + alloc =get_acl (inode,ACL_TYPE_DEFAULT );if (IS_ERR (alloc)) goto fail; + dfacl = alloc;break ;case ACL_TYPE_DEFAULT : - dfacl = acl; - alloc = acl =get_acl (inode,ACL_TYPE_ACCESS ); + alloc =get_acl (inode,ACL_TYPE_ACCESS );if (IS_ERR (alloc)) goto fail; + dfacl = acl; + acl = alloc;break ; } }if (acl ==NULL ) { - alloc = acl =posix_acl_from_mode (inode->i_mode,GFP_KERNEL ); + alloc =posix_acl_from_mode (inode->i_mode,GFP_KERNEL );if (IS_ERR (alloc)) goto fail; + acl = alloc; } status =__nfs3_proc_setacls (inode, acl, dfacl); -posix_acl_release (alloc); +out: +if (acl != orig) +posix_acl_release (acl); +if (dfacl != orig) +posix_acl_release (dfacl);return status; fail: -return PTR_ERR (alloc); + status =PTR_ERR (alloc); + goto out; }const struct xattr_handler *nfs3_xattr_handlers[] = {
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 4/35 vfio: avoid possible overflow in vfio_iommu_type1_pin_pages
In reply to Sasha LevinFrom: Yan Zhao <[email protected]>
[ Upstream commit 0ea971f8dcd6dee78a9a30ea70227cf305f11ff7 ]
add parentheses to avoid possible vaddr overflow.
--- drivers/vfio/vfio_iommu_type1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/vfio/vfio_iommu_type1.c +1 -1
@@ -593,7 +593,7 @@ static int vfio_iommu_type1_pin_pages(void *iommu_data,continue ; } - remote_vaddr = dma->vaddr + iova - dma->iova; + remote_vaddr = dma->vaddr + (iova - dma->iova); ret =vfio_pin_page_external (dma, remote_vaddr, &phys_pfn[i], do_accounting);if (ret)
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 5/35 riscv: fix vdso build with lld
In reply to Sasha LevinFrom: Ilie Halip <[email protected]>
[ Upstream commit 3c1918c8f54166598195d938564072664a8275b1 ]
When building with the LLVM linker this error occurrs:
LD arch/riscv/kernel/vdso/vdso-syms.o
ld.lld: error: no input files
This happens because the lld treats -R as an alias to -rpath, as opposed to ld where -R means --just-symbols.
Use the long option name for compatibility between the two.
--- arch/riscv/kernel/vdso/Makefile | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
arch/riscv/kernel/vdso/Makefile +3 -3
@@ -33,15 +33,15 @@ $(obj)/vdso.so.dbg: $(src)/vdso.lds $(obj-vdso) FORCE $(call if_changed,vdsold) # We also create a special relocatable object that should mirror the symbol -# table and layout of the linked DSO. With ld -R we can then refer to -# these symbols in the kernel code rather than hand-coded addresses. +# table and layout of the linked DSO. With ld --just-symbols we can then +# refer to these symbols in the kernel code rather than hand-coded addresses. SYSCFLAGS_vdso.so.dbg = -shared -s -Wl,-soname=linux-vdso.so.1 \ -Wl,--build-id -Wl,--hash-style=both $(obj)/vdso-dummy.o: $(src)/vdso.lds $(obj)/rt_sigreturn.o FORCE $(call if_changed,vdsold) -LDFLAGS_vdso-syms.o := -r -R +LDFLAGS_vdso-syms.o := -r --just-symbols $(obj)/vdso-syms.o: $(obj)/vdso-dummy.o FORCE $(call if_changed,ld)
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 6/35 scsi: qla2xxx: set UNLOADING before waiting for session deletion
In reply to Sasha LevinFrom: Martin Wilck <[email protected]>
[ Upstream commit 856e152a3c08bf7987cbd41900741d83d9cddc8e ]
The purpose of the UNLOADING flag is to avoid port login procedures to continue when a controller is in the process of shutting down. It makes sense to set this flag before starting session teardown.
Furthermore, use atomic test_and_set_bit() to avoid the shutdown being run multiple times in parallel. In qla2x00_disable_board_on_pci_error(), the test for UNLOADING is postponed until after the check for an already disabled PCI board.
--- drivers/scsi/qla2xxx/qla_os.c | 32 ++++++++++++++------------------ 1 file changed, 14 insertions(+), 18 deletions(-)
drivers/scsi/qla2xxx/qla_os.c +14 -18
@@ -3700,6 +3700,13 @@ qla2x00_remove_one(struct pci_dev *pdev) }qla2x00_wait_for_hba_ready (base_vha); +/* +* if UNLOADING flag is already set, then continue unload, +* where it was set first. +*/ +if (test_and_set_bit (UNLOADING , &base_vha->dpc_flags)) +return ; +if (IS_QLA25XX (ha) ||IS_QLA2031 (ha) ||IS_QLA27XX (ha) ||IS_QLA28XX (ha)) {if (ha->flags.fw_started)
@@ -3718,15 +3725,6 @@ qla2x00_remove_one(struct pci_dev *pdev)qla2x00_wait_for_sess_deletion (base_vha); -/* -* if UNLOAD flag is already set, then continue unload, -* where it was set first. -*/ -if (test_bit (UNLOADING , &base_vha->dpc_flags)) -return ; - -set_bit (UNLOADING , &base_vha->dpc_flags); -qla_nvme_delete (base_vha); dma_free_coherent(&ha->pdev->dev,
@@ -6053,13 +6051,6 @@ qla2x00_disable_board_on_pci_error(struct work_struct *work)struct pci_dev *pdev = ha->pdev;scsi_qla_host_t *base_vha =pci_get_drvdata (ha->pdev); -/* -* if UNLOAD flag is already set, then continue unload, -* where it was set first. -*/ -if (test_bit (UNLOADING , &base_vha->dpc_flags)) -return ; -ql_log (ql_log_warn, base_vha,0x015b ,"Disabling adapter.\n" );
@@ -6070,9 +6061,14 @@ qla2x00_disable_board_on_pci_error(struct work_struct *work)return ; } -qla2x00_wait_for_sess_deletion (base_vha); +/* +* if UNLOADING flag is already set, then continue unload, +* where it was set first. +*/ +if (test_and_set_bit (UNLOADING , &base_vha->dpc_flags)) +return ; -set_bit (UNLOADING , &base_vha->dpc_flags); +qla2x00_wait_for_sess_deletion (base_vha);qla2x00_delete_all_vps (ha, base_vha);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 7/35 scsi: qla2xxx: check UNLOADING before posting async work
In reply to Sasha LevinFrom: Martin Wilck <[email protected]>
[ Upstream commit 5a263892d7d0b4fe351363f8d1a14c6a75955475 ]
qlt_free_session_done() tries to post async PRLO / LOGO, and waits for the completion of these async commands. If UNLOADING is set, this is doomed to timeout, because the async logout command will never complete.
The only way to avoid waiting pointlessly is to fail posting these commands in the first place if the driver is in UNLOADING state. In general, posting any command should be avoided when the driver is UNLOADING.
With this patch, "rmmod qla2xxx" completes without noticeable delay.
--- drivers/scsi/qla2xxx/qla_os.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/scsi/qla2xxx/qla_os.c +3 -0
@@ -4857,6 +4857,9 @@ qla2x00_alloc_work(struct scsi_qla_host *vha, enum qla_work_type type)struct qla_work_evt *e;uint8_t bail; +if (test_bit (UNLOADING , &vha->dpc_flags)) +return NULL ; +QLA_VHA_MARK_BUSY (vha, bail);if (bail)return NULL ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 8/35 scsi: target/iblock: fix WRITE SAME zeroing
In reply to Sasha LevinFrom: David Disseldorp <[email protected]>
[ Upstream commit 1d2ff149b263c9325875726a7804a0c75ef7112e ]
SBC4 specifies that WRITE SAME requests with the UNMAP bit set to zero "shall perform the specified write operation to each LBA specified by the command". Commit 2237498f0b5c ("target/iblock: Convert WRITE_SAME to blkdev_issue_zeroout") modified the iblock backend to call blkdev_issue_zeroout() when handling WRITE SAME requests with UNMAP=0 and a zero data segment.
The iblock blkdev_issue_zeroout() call incorrectly provides a flags parameter of 0 (bool false), instead of BLKDEV_ZERO_NOUNMAP. The bool false parameter reflects the blkdev_issue_zeroout() API prior to commit ee472d835c26 ("block: add a flags argument to (__)blkdev_issue_zeroout") which was merged shortly before 2237498f0b5c.
--- drivers/target/target_core_iblock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/target/target_core_iblock.c +1 -1
@@ -432,7 +432,7 @@ iblock_execute_zero_out(struct block_device *bdev, struct se_cmd *cmd)target_to_linux_sector (dev, cmd->t_task_lba),target_to_linux_sector (dev,sbc_get_write_same_sectors (cmd)), - GFP_KERNEL, false); + GFP_KERNEL, BLKDEV_ZERO_NOUNMAP);if (ret)return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 9/35 RDMA/mlx5: Set GRH fields in query QP on RoCE
In reply to Sasha LevinFrom: Aharon Landau <[email protected]>
[ Upstream commit 2d7e3ff7b6f2c614eb21d0dc348957a47eaffb57 ]
GRH fields such as sgid_index, hop limit, et. are set in the QP context when QP is created/modified.
Currently, when query QP is performed, we fill the GRH fields only if the GRH bit is set in the QP context, but this bit is not set for RoCE. Adjust the check so we will set all relevant data for the RoCE too.
Since this data is returned to userspace, the below is an ABI regression.
--- drivers/infiniband/hw/mlx5/qp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/infiniband/hw/mlx5/qp.c +3 -1
@@ -5496,7 +5496,9 @@ static void to_rdma_ah_attr(struct mlx5_ib_dev *ibdev,rdma_ah_set_path_bits (ah_attr, path->grh_mlid &0x7f );rdma_ah_set_static_rate (ah_attr, path->static_rate ? path->static_rate -5 :0 ); -if (path->grh_mlid & (1 <<7 )) { + +if (path->grh_mlid & (1 <<7 ) || + ah_attr->type ==RDMA_AH_ATTR_TYPE_ROCE ) {u32 tc_fl =be32_to_cpu (path->tclass_flowlabel); rdma_ah_set_grh(ah_attr, NULL,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 10/35 RDMA/core: Prevent mixed use of FDs between shared ufiles
In reply to Sasha LevinFrom: Leon Romanovsky <[email protected]>
[ Upstream commit 0fb00941dc63990a10951146df216fc7b0e20bc2 ]
FDs can only be used on the ufile that created them, they cannot be mixed to other ufiles. We are lacking a check to prevent it.
BUG: KASAN: null-ptr-deref in atomic64_sub_and_test include/asm-generic/atomic-instrumented.h:1547 [inline] BUG: KASAN: null-ptr-deref in atomic_long_sub_and_test include/asm-generic/atomic-long.h:460 [inline] BUG: KASAN: null-ptr-deref in fput_many+0x1a/0x140 fs/file_table.c:336 Write of size 8 at addr 0000000000000038 by task syz-executor179/284 CPU: 0 PID: 284 Comm: syz-executor179 Not tainted 5.5.0-rc5+ #1 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:77 [inline] dump_stack+0x94/0xce lib/dump_stack.c:118 __kasan_report+0x18f/0x1b7 mm/kasan/report.c:510 kasan_report+0xe/0x20 mm/kasan/common.c:639 check_memory_region_inline mm/kasan/generic.c:185 [inline] check_memory_region+0x15d/0x1b0 mm/kasan/generic.c:192 atomic64_sub_and_test include/asm-generic/atomic-instrumented.h:1547 [inline] atomic_long_sub_and_test include/asm-generic/atomic-long.h:460 [inline] fput_many+0x1a/0x140 fs/file_table.c:336 rdma_lookup_put_uobject+0x85/0x130 drivers/infiniband/core/rdma_core.c:692 uobj_put_read include/rdma/uverbs_std_types.h:96 [inline] _ib_uverbs_lookup_comp_file drivers/infiniband/core/uverbs_cmd.c:198 [inline] create_cq+0x375/0xba0 drivers/infiniband/core/uverbs_cmd.c:1006 ib_uverbs_create_cq+0x114/0x140 drivers/infiniband/core/uverbs_cmd.c:1089 ib_uverbs_write+0xaa5/0xdf0 drivers/infiniband/core/uverbs_main.c:769 __vfs_write+0x7c/0x100 fs/read_write.c:494 vfs_write+0x168/0x4a0 fs/read_write.c:558 ksys_write+0xc8/0x200 fs/read_write.c:611 do_syscall_64+0x9c/0x390 arch/x86/entry/common.c:294 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x44ef99 Code: 00 b8 00 01 00 00 eb e1 e8 74 1c 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 c4 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007ffc0b74c028 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 00007ffc0b74c030 RCX: 000000000044ef99 RDX: 0000000000000040 RSI: 0000000020000040 RDI: 0000000000000005 RBP: 00007ffc0b74c038 R08: 0000000000401830 R09: 0000000000401830 R10: 00007ffc0b74c038 R11: 0000000000000246 R12: 0000000000000000 R13: 0000000000000000 R14: 00000000006be018 R15: 0000000000000000
--- drivers/infiniband/core/rdma_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/core/rdma_core.c +1 -1
@@ -362,7 +362,7 @@ lookup_get_fd_uobject(const struct uverbs_api_object *obj,* and the caller is expected to ensure that uverbs_close_fd is never * done while a call top lookup is possible. */ -if (f->f_op != fd_type->fops) { +if (f->f_op != fd_type->fops || uobject->ufile != ufile) {fput (f);return ERR_PTR (-EBADF ); }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 11/35 dmaengine: pch_dma.c: Avoid data race between probe and irq handler
In reply to Sasha LevinFrom: Madhuparna Bhowmik <[email protected]>
[ Upstream commit 2e45676a4d33af47259fa186ea039122ce263ba9 ]
pd->dma.dev is read in irq handler pd_irq().
However, it is set to pdev->dev after request_irq(). Therefore, set pd->dma.dev to pdev->dev before request_irq() to avoid data race between pch_dma_probe() and pd_irq().
Found by Linux Driver Verification project (linuxtesting.org).
--- drivers/dma/pch_dma.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/dma/pch_dma.c +1 -1
@@ -865,6 +865,7 @@ static int pch_dma_probe(struct pci_dev *pdev, }pci_set_master (pdev); + pd->dma.dev = &pdev->dev; err =request_irq (pdev->irq, pd_irq,IRQF_SHARED ,DRV_NAME , pd);if (err) {
@@ -880,7 +881,6 @@ static int pch_dma_probe(struct pci_dev *pdev, goto err_free_irq; } - pd->dma.dev = &pdev->dev;INIT_LIST_HEAD (&pd->dma.channels);
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 12/35 dmaengine: mmp_tdma: Do not ignore slave config validation errors
In reply to Sasha LevinFrom: Lubomir Rintel <[email protected]>
[ Upstream commit 363c32701c7fdc8265a84b21a6a4f45d1202b9ca ]
With an invalid dma_slave_config set previously, mmp_tdma_prep_dma_cyclic() would detect an error whilst configuring the channel, but proceed happily on:
[ 120.756530] mmp-tdma d42a0800.adma: mmp_tdma: unknown burst size.
--- drivers/dma/mmp_tdma.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/dma/mmp_tdma.c +2 -1
@@ -443,7 +443,8 @@ static struct dma_async_tx_descriptor *mmp_tdma_prep_dma_cyclic(if (!desc) goto err_out; -mmp_tdma_config_write (chan, direction, &tdmac->slave_config); +if (mmp_tdma_config_write (chan, direction, &tdmac->slave_config)) + goto err_out;while (buf < buf_len) { desc = &tdmac->desc_arr[i];
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 13/35 dmaengine: mmp_tdma: Reset channel error on release
In reply to Sasha LevinFrom: Lubomir Rintel <[email protected]>
[ Upstream commit 0c89446379218698189a47871336cb30286a7197 ]
When a channel configuration fails, the status of the channel is set to DEV_ERROR so that an attempt to submit it fails. However, this status sticks until the heat end of the universe, making it impossible to recover from the error.
Let's reset it when the channel is released so that further use of the channel with correct configuration is not impacted.
--- drivers/dma/mmp_tdma.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/dma/mmp_tdma.c +2 -0
@@ -363,6 +363,8 @@ static void mmp_tdma_free_descriptor(struct mmp_tdma_chan *tdmac)gen_pool_free (gpool, (unsigned long )tdmac->desc_arr, size); tdmac->desc_arr =NULL ; +if (tdmac->status ==DMA_ERROR ) + tdmac->status =DMA_COMPLETE ;return ; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 14/35 vfio/type1: Fix VA->PA translation for PFNMAP VMAs in vaddr_get_pfn()
In reply to Sasha LevinFrom: Sean Christopherson <[email protected]>
[ Upstream commit 5cbf3264bc715e9eb384e2b68601f8c02bb9a61d ]
Use follow_pfn() to get the PFN of a PFNMAP VMA instead of assuming that vma->vm_pgoff holds the base PFN of the VMA. This fixes a bug where attempting to do VFIO_IOMMU_MAP_DMA on an arbitrary PFNMAP'd region of memory calculates garbage for the PFN.
Hilariously, this only got detected because the first "PFN" calculated by vaddr_get_pfn() is PFN 0 (vma->vm_pgoff==0), and iommu_iova_to_phys() uses PA==0 as an error, which triggers a WARN in vfio_unmap_unpin() because the translation "failed". PFN 0 is now unconditionally reserved on x86 in order to mitigate L1TF, which causes is_invalid_reserved_pfn() to return true and in turns results in vaddr_get_pfn() returning success for PFN 0. Eventually the bogus calculation runs into PFNs that aren't reserved and leads to failure in vfio_pin_map_dma(). The subsequent call to vfio_remove_dma() attempts to unmap PFN 0 and WARNs.
WARNING: CPU: 8 PID: 5130 at drivers/vfio/vfio_iommu_type1.c:750 vfio_unmap_unpin+0x2e1/0x310 [vfio_iommu_type1] Modules linked in: vfio_pci vfio_virqfd vfio_iommu_type1 vfio ... CPU: 8 PID: 5130 Comm: sgx Tainted: G W 5.6.0-rc5-705d787c7fee-vfio+ #3 Hardware name: Intel Corporation Mehlow UP Server Platform/Moss Beach Server, BIOS CNLSE2R1.D00.X119.B49.1803010910 03/01/2018 RIP: 0010:vfio_unmap_unpin+0x2e1/0x310 [vfio_iommu_type1] Code: <0f> 0b 49 81 c5 00 10 00 00 e9 c5 fe ff ff bb 00 10 00 00 e9 3d fe RSP: 0018:ffffbeb5039ebda8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff9a55cbf8d480 RCX: 0000000000000000 RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff9a52b771c200 RBP: 0000000000000000 R08: 0000000000000040 R09: 00000000fffffff2 R10: 0000000000000001 R11: ffff9a51fa896000 R12: 0000000184010000 R13: 0000000184000000 R14: 0000000000010000 R15: ffff9a55cb66ea08 FS: 00007f15d3830b40(0000) GS:ffff9a55d5600000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000561cf39429e0 CR3: 000000084f75f005 CR4: 00000000003626e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: vfio_remove_dma+0x17/0x70 [vfio_iommu_type1] vfio_iommu_type1_ioctl+0x9e3/0xa7b [vfio_iommu_type1] ksys_ioctl+0x92/0xb0 __x64_sys_ioctl+0x16/0x20 do_syscall_64+0x4c/0x180 entry_SYSCALL_64_after_hwframe+0x44/0xa9 RIP: 0033:0x7f15d04c75d7 Code: <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 81 48 2d 00 f7 d8 64 89 01 48
--- drivers/vfio/vfio_iommu_type1.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/vfio/vfio_iommu_type1.c +2 -2
@@ -380,8 +380,8 @@ static int vaddr_get_pfn(struct mm_struct *mm, unsigned long vaddr, vma =find_vma_intersection (mm, vaddr, vaddr +1 );if (vma && vma->vm_flags &VM_PFNMAP ) { - *pfn = ((vaddr - vma->vm_start) >>PAGE_SHIFT ) + vma->vm_pgoff; -if (is_invalid_reserved_pfn (*pfn)) +if (!follow_pfn (vma, vaddr, pfn) && +is_invalid_reserved_pfn (*pfn)) ret =0 ; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 15/35 ALSA: hda: Match both PCI ID and SSID for driver blacklist
In reply to Sasha LevinFrom: Takashi Iwai <[email protected]>
[ Upstream commit 977dfef40c8996b69afe23a9094d184049efb7bb ]
The commit 3c6fd1f07ed0 ("ALSA: hda: Add driver blacklist") added a new blacklist for the devices that are known to have empty codecs, and one of the entries was ASUS ROG Zenith II (PCI SSID 1043:874f). However, it turned out that the very same PCI SSID is used for the previous model that does have the valid HD-audio codecs and the change broke the sound on it.
Since the empty codec problem appear on the certain AMD platform (PCI ID 1022:1487), this patch changes the blacklist matching to both PCI ID and SSID using pci_match_id(). Also, the entry that was removed by the previous fix for ASUS ROG Zenigh II is re-added.
--- sound/pci/hda/hda_intel.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
sound/pci/hda/hda_intel.c +5 -4
@@ -2023,9 +2023,10 @@ static void pcm_mmap_prepare(struct snd_pcm_substream *substream,* some HD-audio PCI entries are exposed without any codecs, and such devices * should be ignored from the beginning. */ -static const struct snd_pci_quirk driver_blacklist[] = { -SND_PCI_QUIRK (0x1462 ,0xcb59 ,"MSI TRX40 Creator" ,0 ), -SND_PCI_QUIRK (0x1462 ,0xcb60 ,"MSI TRX40" ,0 ), +static const struct pci_device_id driver_blacklist[] = { + {PCI_DEVICE_SUB (0x1022 ,0x1487 ,0x1043 ,0x874f ) },/* ASUS ROG Zenith II / Strix */ + {PCI_DEVICE_SUB (0x1022 ,0x1487 ,0x1462 ,0xcb59 ) },/* MSI TRX40 Creator */ + {PCI_DEVICE_SUB (0x1022 ,0x1487 ,0x1462 ,0xcb60 ) },/* MSI TRX40 */ {} };
@@ -2064,7 +2065,7 @@ static int azx_probe(struct pci_dev *pci,bool schedule_probe;int err; -if (snd_pci_quirk_lookup (pci, driver_blacklist)) { +if (pci_match_id (driver_blacklist, pci)) {dev_info (&pci->dev,"Skipping the blacklisted device\n" );return -ENODEV ; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 16/35 selftests/ftrace: Check the first record for kprobe_args_type.tc
In reply to Sasha LevinFrom: Xiao Yang <[email protected]>
[ Upstream commit f0c0d0cf590f71b2213b29a7ded2cde3d0a1a0ba ]
It is possible to get multiple records from trace during test and then more
than 4 arguments are assigned to ARGS. This situation results in the failure
of kprobe_args_type.tc. For example:
-----------------------------------------------------------
grep testprobe trace
ftracetest-5902 [001] d... 111195.682227: testprobe: (_do_fork+0x0/0x460) arg1=334823024 arg2=334823024 arg3=0x13f4fe70 arg4=7
pmlogger-5949 [000] d... 111195.709898: testprobe: (_do_fork+0x0/0x460) arg1=345308784 arg2=345308784 arg3=0x1494fe70 arg4=7
grep testprobe trace
sed -e 's/.* arg1=\(.*\) arg2=\(.*\) arg3=\(.*\) arg4=\(.*\)/\1 \2 \3 \4/'
ARGS='334823024 334823024 0x13f4fe70 7
345308784 345308784 0x1494fe70 7'
-----------------------------------------------------------
We don't care which process calls do_fork so just check the first record to fix the issue.
--- .../testing/selftests/ftrace/test.d/kprobe/kprobe_args_type.tc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/ftrace/test.d/kprobe/kprobe_args_type.tc +1 -1
@@ -38,7 +38,7 @@ for width in 64 32 16 8; do echo 0 > events/kprobes/testprobe/enable : "Confirm the arguments is recorded in given types correctly" - ARGS=`grep "testprobe" trace | sed -e 's/.* arg1=\(.*\) arg2=\(.*\) arg3=\(.*\) arg4=\(.*\)/\1 \2 \3 \4/'` + ARGS=`grep "testprobe" trace | head -n 1 | sed -e 's/.* arg1=\(.*\) arg2=\(.*\) arg3=\(.*\) arg4=\(.*\)/\1 \2 \3 \4/'` check_types $ARGS $width : "Clear event for next loop"
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 17/35 RDMA/core: Fix race between destroy and release FD object
In reply to Sasha LevinFrom: Leon Romanovsky <[email protected]>
[ Upstream commit f0abc761bbb9418876cc4d1ebc473e4ea6352e42 ]
The call to ->lookup_put() was too early and it caused an unlock of the read/write protection of the uobject after the FD was put. This allows a race:
CPU1 CPU2
rdma_lookup_put_uobject()
lookup_put_fd_uobject()
fput() fput() uverbs_uobject_fd_release() WARN_ON(uverbs_try_lock_object(uobj, UVERBS_LOOKUP_WRITE));
atomic_dec(usecnt)
Fix the code by changing the order, first unlock and call to ->lookup_put() after that.
--- drivers/infiniband/core/rdma_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/core/rdma_core.c +1 -1
@@ -689,7 +689,6 @@ void rdma_lookup_put_uobject(struct ib_uobject *uobj, enum rdma_lookup_mode mode) {assert_uverbs_usecnt (uobj, mode); - uobj->uapi_object->type_class->lookup_put (uobj, mode);/* * In order to unlock an object, either decrease its usecnt for * read access or zero it in case of exclusive access. See
@@ -706,6 +705,7 @@ void rdma_lookup_put_uobject(struct ib_uobject *uobj,break ; } + uobj->uapi_object->type_class->lookup_put (uobj, mode);/* Pairs with the kref obtained by type->lookup_get */ uverbs_uobject_put (uobj); }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 18/35 cpufreq: intel_pstate: Only mention the BIOS disabling turbo mode once
In reply to Sasha LevinFrom: Chris Wilson <[email protected]>
[ Upstream commit 8c539776ac83c0857395e1ccc9c6b516521a2d32 ]
Make a note of the first time we discover the turbo mode has been disabled by the BIOS, as otherwise we complain every time we try to update the mode.
--- drivers/cpufreq/intel_pstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/cpufreq/intel_pstate.c +1 -1
@@ -1058,7 +1058,7 @@ static ssize_t store_no_turbo(struct kobject *a, struct kobj_attribute *b,update_turbo_state ();if (global.turbo_disabled) { -pr_warn ("Turbo disabled by BIOS or unavailable on processor\n" ); +pr_notice_once ("Turbo disabled by BIOS or unavailable on processor\n" );mutex_unlock (&intel_pstate_limits_lock);mutex_unlock (&intel_pstate_driver_lock);return -EPERM ;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 19/35 dma-buf: Fix SET_NAME ioctl uapi
In reply to Sasha LevinFrom: Daniel Vetter <[email protected]>
[ Upstream commit a5bff92eaac45bdf6221badf9505c26792fdf99e ]
The uapi is the same on 32 and 64 bit, but the number isn't. Everyone who botched this please re-read:
https://www.kernel.org/doc/html/v5.4-preprc-cpu/ioctl/botching-up-ioctls.html
Also, the type argument for the ioctl macros is for the type the void __user *arg pointer points at, which in this case would be the variable-sized char[] of a 0 terminated string. So this was botched in more than just the usual ways.
--- drivers/dma-buf/dma-buf.c | 3 ++- include/uapi/linux/dma-buf.h | 6 ++++++ 2 files changed, 8 insertions(+), 1 deletion(-)
drivers/dma-buf/dma-buf.c +2 -1
@@ -388,7 +388,8 @@ static long dma_buf_ioctl(struct file *file,return ret; -case DMA_BUF_SET_NAME : +case DMA_BUF_SET_NAME_A : +case DMA_BUF_SET_NAME_B :return dma_buf_set_name (dmabuf, (const char __user *)arg);default :
include/uapi/linux/dma-buf.h +6 -0
@@ -39,6 +39,12 @@ struct dma_buf_sync {#define DMA_BUF_BASE 'b'#define DMA_BUF_IOCTL_SYNC _IOW(DMA_BUF_BASE, 0, struct dma_buf_sync) + +/* 32/64bitness of this uapi was botched in android, there's no difference +* between them in actual uapi, they're just different numbers. +*/ #define DMA_BUF_SET_NAME _IOW(DMA_BUF_BASE, 1, const char *) +#define DMA_BUF_SET_NAME_A _IOW(DMA_BUF_BASE, 1, u32) +#define DMA_BUF_SET_NAME_B _IOW(DMA_BUF_BASE, 1, u64)#endif
Signature · 6 lines
2.20.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 20/35 nvme: prevent double free in nvme_alloc_ns() error handling
In reply to Sasha LevinFrom: Niklas Cassel <[email protected]>
[ Upstream commit 132be62387c7a72a38872676c18b0dfae264adb8 ]
When jumping to the out_put_disk label, we will call put_disk(), which will trigger a call to disk_release(), which calls blk_put_queue().
Later in the cleanup code, we do blk_cleanup_queue(), which will also call blk_put_queue().
Putting the queue twice is incorrect, and will generate a KASAN splat.
Set the disk->queue pointer to NULL, before calling put_disk(), so that the first call to blk_put_queue() will not free the queue.
The second call to blk_put_queue() uses another pointer to the same queue, so this call will still free the queue.
--- drivers/nvme/host/core.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/nvme/host/core.c +2 -0
@@ -3566,6 +3566,8 @@ static int nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)return 0 ; out_put_disk: +/* prevent double queue cleanup */ + ns->disk->queue =NULL ;put_disk (ns->disk); out_unlink_ns:mutex_lock (&ctrl->subsys->lock);
Signature · 7 lines
2.20.1 _______________________________________________ linux-nvme mailing list [email protected] http://lists.infradead.org/mailman/listinfo/linux-nvme
Sasha Levin
maintainerPATCH 21/35 dmaengine: dmatest: Fix iteration non-stop logic
In reply to Sasha LevinFrom: Andy Shevchenko <[email protected]>
[ Upstream commit b9f960201249f20deea586b4ec814669b4c6b1c0 ]
Under some circumstances, i.e. when test is still running and about to time out and user runs, for example,
grep -H . /sys/module/dmatest/parameters/*
the iterations parameter is not respected and test is going on and on until user gives
echo 0 > /sys/module/dmatest/parameters/run
This is not what expected.
The history of this bug is interesting. I though that the commit
2d88ce76eb98 ("dmatest: add a 'wait' parameter")
is a culprit, but looking closer to the code I think it simple revealed the broken logic from the day one, i.e. in the commit
0a2ff57d6fba ("dmaengine: dmatest: add a maximum number of test iterations")
which adds iterations parameter.
So, to the point, the conditional of checking the thread to be stopped being first part of conjunction logic prevents to check iterations. Thus, we have to always check both conditions to be able to stop after given iterations.
Since it wasn't visible before second commit appeared, I add a respective Fixes tag.
--- drivers/dma/dmatest.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/dma/dmatest.c +2 -2
@@ -662,8 +662,8 @@ static int dmatest_func(void *data) flags =DMA_CTRL_ACK |DMA_PREP_INTERRUPT ; ktime =ktime_get (); -while (!kthread_should_stop () - && !(params->iterations && total_tests >= params->iterations)) { +while (!(kthread_should_stop () || + (params->iterations && total_tests >= params->iterations))) {struct dma_async_tx_descriptor *tx =NULL ;struct dmaengine_unmap_data *um;dma_addr_t *dsts;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 22/35 i2c: iproc: generate stop event for slave writes
In reply to Sasha LevinFrom: Rayagonda Kokatanur <[email protected]>
[ Upstream commit 068143a8195fb0fdeea1f3ca430b3db0f6d04a53 ]
When slave status is I2C_SLAVE_RX_END, generate I2C_SLAVE_STOP event to i2c_client.
--- drivers/i2c/busses/i2c-bcm-iproc.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/i2c/busses/i2c-bcm-iproc.c +3 -0
@@ -359,6 +359,9 @@ static bool bcm_iproc_i2c_slave_isr(struct bcm_iproc_i2c_dev *iproc_i2c, value = (u8)((val >>S_RX_DATA_SHIFT ) &S_RX_DATA_MASK );i2c_slave_event (iproc_i2c->slave,I2C_SLAVE_WRITE_RECEIVED , &value); +if (rx_status ==I2C_SLAVE_RX_END ) +i2c_slave_event (iproc_i2c->slave, +I2C_SLAVE_STOP , &value); } }else if (status &BIT (IS_S_TX_UNDERRUN_SHIFT )) {/* Master read other than start */
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 23/35 ALSA: hda/hdmi: fix race in monitor detection during probe
In reply to Sasha LevinFrom: Kai Vehmanen <[email protected]>
[ Upstream commit ca76282b6faffc83601c25bd2a95f635c03503ef ]
A race exists between build_pcms() and build_controls() phases of codec setup. Build_pcms() sets up notifier for jack events. If a monitor event is received before build_controls() is run, the initial jack state is lost and never reported via mixer controls.
The problem can be hit at least with SOF as the controller driver. SOF calls snd_hda_codec_build_controls() in its workqueue-based probe and this can be delayed enough to hit the race condition.
Fix the issue by invalidating the per-pin ELD information when build_controls() is called. The existing call to hdmi_present_sense() will update the ELD contents. This ensures initial monitor state is correctly reflected via mixer controls.
--- sound/pci/hda/patch_hdmi.c | 2 ++ 1 file changed, 2 insertions(+)
sound/pci/hda/patch_hdmi.c +2 -0
@@ -2232,7 +2232,9 @@ static int generic_hdmi_build_controls(struct hda_codec *codec)for (pin_idx =0 ; pin_idx < spec->num_pins; pin_idx++) {struct hdmi_spec_per_pin *per_pin =get_pin (spec, pin_idx); +struct hdmi_eld *pin_eld = &per_pin->sink_eld; + pin_eld->eld_valid = false;hdmi_present_sense (per_pin,0 ); }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 24/35 dmaengine: dmatest: Fix process hang when reading 'wait' parameter
In reply to Sasha LevinFrom: Andy Shevchenko <[email protected]>
[ Upstream commit aa72f1d20ee973d68f26d46fce5e1cf6f9b7e1ca ]
If we do
% echo 1 > /sys/module/dmatest/parameters/run
[ 115.851124] dmatest: Could not start test, no channels configured
% echo dma8chan7 > /sys/module/dmatest/parameters/channel
[ 127.563872] dmatest: Added 1 threads using dma8chan7
% cat /sys/module/dmatest/parameters/wait
... !!! HANG !!! ...
The culprit is the commit 6138f967bccc
("dmaengine: dmatest: Use fixed point div to calculate iops")
which makes threads not to run, but pending and being kicked off by writing to the 'run' node. However, it forgot to consider 'wait' routine to avoid above mentioned case.
In order to fix this, check for really running threads, i.e. with pending and done flags unset.
It's pity the culprit commit hadn't updated documentation and tested all possible scenarios.
--- drivers/dma/dmatest.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/dma/dmatest.c +1 -1
@@ -240,7 +240,7 @@ static bool is_threaded_test_run(struct dmatest_info *info)struct dmatest_thread *thread;list_for_each_entry (thread, &dtc->threads, node) { -if (!thread->done) +if (!thread->done && !thread->pending)return true; } }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 25/35 drm/amd/powerplay: avoid using pm_en before it is initialized revised
In reply to Sasha LevinFrom: Tiecheng Zhou <[email protected]>
[ Upstream commit 690ae30be163d5262feae01335b2a6f30569e5aa ]
hwmgr->pm_en is initialized at hwmgr_hw_init.
during amdgpu_device_init, there is amdgpu_asic_reset that calls to soc15_asic_reset (for V320 usecase, Vega10 asic), in which:
1) soc15_asic_reset_method calls to pp_get_asic_baco_capability (pm_en)
2) soc15_asic_baco_reset calls to pp_set_asic_baco_state (pm_en)
pm_en is used in the above two cases while it has not yet been initialized
So avoid using pm_en in the above two functions for V320 passthrough.
--- drivers/gpu/drm/amd/powerplay/amd_powerplay.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/powerplay/amd_powerplay.c +4 -2
@@ -1425,7 +1425,8 @@ static int pp_get_asic_baco_capability(void *handle, bool *cap)if (!hwmgr)return -EINVAL ; -if (!hwmgr->pm_en || !hwmgr->hwmgr_func->get_asic_baco_capability) +if (!(hwmgr->not_vf && amdgpu_dpm) || + !hwmgr->hwmgr_func->get_asic_baco_capability)return 0 ;mutex_lock (&hwmgr->smu_lock);
@@ -1459,7 +1460,8 @@ static int pp_set_asic_baco_state(void *handle, int state)if (!hwmgr)return -EINVAL ; -if (!hwmgr->pm_en || !hwmgr->hwmgr_func->set_asic_baco_state) +if (!(hwmgr->not_vf && amdgpu_dpm) || + !hwmgr->hwmgr_func->set_asic_baco_state)return 0 ;mutex_lock (&hwmgr->smu_lock);
Signature · 6 lines
2.20.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 26/35 SUNRPC: defer slow parts of rpc_free_client() to a workqueue.
In reply to Sasha LevinFrom: NeilBrown <[email protected]>
[ Upstream commit 7c4310ff56422ea43418305d22bbc5fe19150ec4 ]
The rpciod workqueue is on the write-out path for freeing dirty memory, so it is important that it never block waiting for memory to be allocated - this can lead to a deadlock.
rpc_execute() - which is often called by an rpciod work item - calls rcp_task_release_client() which can lead to rpc_free_client().
rpc_free_client() makes two calls which could potentially block wating for memory allocation.
rpc_clnt_debugfs_unregister() calls into debugfs and will block while any of the debugfs files are being accessed. In particular it can block while any of the 'open' methods are being called and all of these use malloc for one thing or another. So this can deadlock if the memory allocation waits for NFS to complete some writes via rpciod.
rpc_clnt_remove_pipedir() can take the inode_lock() and while it isn't obvious that memory allocations can happen while the lock it held, it is safer to assume they might and to not let rpciod call rpc_clnt_remove_pipedir().
So this patch moves these two calls (together with the final kfree() and rpciod_down()) into a work-item to be run from the system work-queue. rpciod can continue its important work, and the final stages of the free can happen whenever they happen.
I have seen this deadlock on a 4.12 based kernel where debugfs used synchronize_srcu() when removing objects. synchronize_srcu() requires a workqueue and there were no free workther threads and none could be allocated. While debugsfs no longer uses SRCU, I believe the deadlock is still possible.
--- include/linux/sunrpc/clnt.h | 8 +++++++- net/sunrpc/clnt.c | 21 +++++++++++++++++---- 2 files changed, 24 insertions(+), 5 deletions(-)
include/linux/sunrpc/clnt.h +7 -1
@@ -71,7 +71,13 @@ struct rpc_clnt {#if IS_ENABLED (CONFIG_SUNRPC_DEBUG )struct dentry *cl_debugfs;/* debugfs directory */ #endif -struct rpc_xprt_iter cl_xpi; +/* cl_work is only needed after cl_xpi is no longer used, +* and that are of similar size +*/ +union { +struct rpc_xprt_iter cl_xpi; +struct work_struct cl_work; + };const struct cred *cl_cred; };
net/sunrpc/clnt.c +17 -4
@@ -877,6 +877,20 @@ EXPORT_SYMBOL_GPL(rpc_shutdown_client);/* * Free an RPC client */ +static void rpc_free_client_work (struct work_struct *work) +{ + struct rpc_clnt *clnt = container_of(work, struct rpc_clnt, cl_work); + +/* These might block on processes that might allocate memory, +* so they cannot be called in rpciod, so they are handled separately +* here. +*/ +rpc_clnt_debugfs_unregister (clnt); +rpc_clnt_remove_pipedir (clnt); + +kfree (clnt); +rpciod_down (); +}static struct rpc_clnt *rpc_free_client (struct rpc_clnt *clnt) {
@@ -887,17 +901,16 @@ rpc_free_client(struct rpc_clnt *clnt) rcu_dereference(clnt->cl_xprt)->servername);if (clnt->cl_parent != clnt) parent = clnt->cl_parent; -rpc_clnt_debugfs_unregister (clnt); -rpc_clnt_remove_pipedir (clnt);rpc_unregister_client (clnt);rpc_free_iostats (clnt->cl_metrics); clnt->cl_metrics =NULL ;xprt_put (rcu_dereference_raw (clnt->cl_xprt));xprt_iter_destroy (&clnt->cl_xpi); -rpciod_down ();put_cred (clnt->cl_cred);rpc_free_clid (clnt); -kfree (clnt); + +INIT_WORK (&clnt->cl_work, rpc_free_client_work); +schedule_work (&clnt->cl_work);return parent; }
Signature · 1 line
2.20.1
Re: [PATCH AUTOSEL 5.4 26/35] SUNRPC: defer slow parts of rpc_free_client() to a workqueue.
Quoted 3 lines · Sasha Levin
From: NeilBrown <[email protected]>
[ Upstream commit
7c4310ff56422ea43418305d22bbc5fe19150ec4]
This one is buggy - it introduces a use-after-free. Best delay it for now.
NeilBrown
Quoted 108 lines
The rpciod workqueue is on the write-out path for freeing dirty memory, so it is important that it never block waiting for memory to be allocated - this can lead to a deadlock.
rpc_execute()- which is often called by an rpciod work item - callsrcp_task_release_client()which can lead torpc_free_client().
rpc_free_client()makes two calls which could potentially block wating for memory allocation.
rpc_clnt_debugfs_unregister()calls into debugfs and will block while any of the debugfs files are being accessed. In particular it can block while any of the 'open' methods are being called and all of these use malloc for one thing or another. So this can deadlock if the memory allocation waits for NFS to complete some writes via rpciod.
rpc_clnt_remove_pipedir()can take theinode_lock()and while it isn't obvious that memory allocations can happen while the lock it held, it is safer to assume they might and to not let rpciod callrpc_clnt_remove_pipedir().So this patch moves these two calls (together with the final
kfree()andrpciod_down()) into a work-item to be run from the system work-queue. rpciod can continue its important work, and the final stages of the free can happen whenever they happen.I have seen this deadlock on a 4.12 based kernel where debugfs used
synchronize_srcu()when removing objects.synchronize_srcu()requires a workqueue and there were no free workther threads and none could be allocated. While debugsfs no longer uses SRCU, I believe the deadlock is still possible.Signed-off-by NeilBrown <[email protected]>Signed-off-by Trond Myklebust <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- include/linux/sunrpc/clnt.h | 8 +++++++- net/sunrpc/clnt.c | 21 +++++++++++++++++---- 2 files changed, 24 insertions(+), 5 deletions(-)include/linux/sunrpc/clnt.h +7
-1@@ -71,7 +71,13 @@ struct rpc_clnt {#if IS_ENABLED (CONFIG_SUNRPC_DEBUG )struct dentry *cl_debugfs;/* debugfs directory */ #endif -struct rpc_xprt_iter cl_xpi; +/* cl_work is only needed after cl_xpi is no longer used, +* and that are of similar size +*/ +union { +struct rpc_xprt_iter cl_xpi; +struct work_struct cl_work; + };const struct cred *cl_cred; };net/sunrpc/clnt.c +17
-4@@ -877,6 +877,20 @@ EXPORT_SYMBOL_GPL(rpc_shutdown_client);/* * Free an RPC client */ +static void rpc_free_client_work (struct work_struct *work) +{ + struct rpc_clnt *clnt = container_of(work, struct rpc_clnt, cl_work); + +/* These might block on processes that might allocate memory, +* so they cannot be called in rpciod, so they are handled separately +* here. +*/ +rpc_clnt_debugfs_unregister (clnt); +rpc_clnt_remove_pipedir (clnt); + +kfree (clnt); +rpciod_down (); +}static struct rpc_clnt *rpc_free_client (struct rpc_clnt *clnt) {@@ -887,17 +901,16 @@ rpc_free_client(struct rpc_clnt *clnt) rcu_dereference(clnt->cl_xprt)->servername);if (clnt->cl_parent != clnt) parent = clnt->cl_parent; -rpc_clnt_debugfs_unregister (clnt); -rpc_clnt_remove_pipedir (clnt);rpc_unregister_client (clnt);rpc_free_iostats (clnt->cl_metrics); clnt->cl_metrics =NULL ;xprt_put (rcu_dereference_raw (clnt->cl_xprt));xprt_iter_destroy (&clnt->cl_xpi); -rpciod_down ();put_cred (clnt->cl_cred);rpc_free_clid (clnt); -kfree (clnt); + +INIT_WORK (&clnt->cl_work, rpc_free_client_work); +schedule_work (&clnt->cl_work);return parent; }Signature · 1 line
2.20.1
Sasha Levin
maintainerRe: [PATCH AUTOSEL 5.4 26/35] SUNRPC: defer slow parts of rpc_free_client() to a workqueue.
Quoted 8 lines · NeilBrown
Quoted 3 lines · Sasha Levin
From: NeilBrown <[email protected]>
[ Upstream commit
7c4310ff56422ea43418305d22bbc5fe19150ec4]This one is buggy - it introduces a use-after-free. Best delay it for now.
I've dropped it, thanks!
Signature · 2 lines
Thanks, Sasha
Sasha Levin
maintainerPATCH 27/35 drm/amd/display: check if REFCLK_CNTL register is present
In reply to Sasha LevinFrom: Dmytro Laktyushkin <[email protected]>
[ Upstream commit 3159d41db3a04330c31ece32f8b29752fc114848 ]
Check before programming the register since it isn't present on all IPs using this code.
--- drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +2 -1
@@ -2015,7 +2015,8 @@ static void dcn20_fpga_init_hw(struct dc *dc)REG_UPDATE (DCHUBBUB_GLOBAL_TIMER_CNTL ,DCHUBBUB_GLOBAL_TIMER_REFDIV ,2 );REG_UPDATE (DCHUBBUB_GLOBAL_TIMER_CNTL ,DCHUBBUB_GLOBAL_TIMER_ENABLE ,1 ); -REG_WRITE (REFCLK_CNTL ,0 ); +if (REG (REFCLK_CNTL )) +REG_WRITE (REFCLK_CNTL ,0 );//
Signature · 6 lines
2.20.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Sasha Levin
maintainerPATCH 28/35 drm/amd/display: Update downspread percent to match spreadsheet for DCN2.1
In reply to Sasha LevinFrom: Sung Lee <[email protected]>
[ Upstream commit 668a6741f809f2d15d125cfe2b39661e8f1655ea ]
[WHY]
The downspread percentage was copied over from a previous version of the display_mode_lib spreadsheet. This value has been updated, and the previous value is too high to allow for such modes as 4K120hz. The new value is sufficient for such modes.
[HOW]
Update the value in dcn21_resource to match the spreadsheet.
--- drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +1 -1
@@ -247,7 +247,7 @@ struct _vcs_dpi_soc_bounding_box_st dcn2_1_soc = { .dram_channel_width_bytes = 4, .fabric_datapath_to_dcn_data_return_bytes = 32, .dcn_downspread_percent = 0.5, - .downspread_percent = 0.5, + .downspread_percent = 0.38, .dram_page_open_time_ns = 50.0, .dram_rw_turnaround_time_ns = 17.5, .dram_return_buffer_per_channel_bytes = 8192,
Signature · 6 lines
2.20.1 _______________________________________________ amd-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/amd-gfx
Sasha Levin
maintainerPATCH 29/35 Fix use after free in get_tree_bdev()
In reply to Sasha LevinFrom: David Howells <[email protected]>
[ Upstream commit dd7bc8158b413e0b580c491e8bd18cb91057c7c2 ]
Commit 6fcf0c72e4b9, a fix to get_tree_bdev() put a missing blkdev_put() in the wrong place, before a warnf() that displays the bdev under consideration rather after it.
This results in a silent lockup in printk("%pg") called via warnf() from get_tree_bdev() under some circumstances when there's a race with the blockdev being frozen. This can be caused by xfstests/tests/generic/085 in combination with Lukas Czerner's ext4 mount API conversion patchset. It looks like it ought to occur with other users of get_tree_bdev() such as XFS, but apparently doesn't.
Fix this by switching the order of the lines.
--- fs/super.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/super.c +1 -1
@@ -1302,8 +1302,8 @@ int get_tree_bdev(struct fs_context *fc,mutex_lock (&bdev->bd_fsfreeze_mutex);if (bdev->bd_fsfreeze_count >0 ) {mutex_unlock (&bdev->bd_fsfreeze_mutex); -blkdev_put (bdev, mode);warnf (fc,"%pg: Can't mount, blockdev is frozen" , bdev); +blkdev_put (bdev, mode);return -EBUSY ; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 30/35 drm/qxl: lost qxl_bo_kunmap_atomic_page in qxl_image_init_helper()
In reply to Sasha LevinFrom: Vasily Averin <[email protected]>
[ Upstream commit 5b5703dbafae74adfbe298a56a81694172caf5e6 ]
v2: removed TODO reminder
--- drivers/gpu/drm/qxl/qxl_image.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/qxl/qxl_image.c +2 -1
@@ -212,7 +212,8 @@ qxl_image_init_helper(struct qxl_device *qdev,break ;default :DRM_ERROR ("unsupported image bit depth\n" ); -return -EINVAL ;/* TODO: cleanup */ +qxl_bo_kunmap_atomic_page (qdev, image_bo, ptr); +return -EINVAL ; } image->u.bitmap.flags =QXL_BITMAP_TOP_DOWN ; image->u.bitmap.x = width;
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 31/35 ALSA: opti9xx: shut up gcc-10 range warning
In reply to Sasha LevinFrom: Arnd Bergmann <[email protected]>
[ Upstream commit 5ce00760a84848d008554c693ceb6286f4d9c509 ]
gcc-10 points out a few instances of suspicious integer arithmetic leading to value truncation:
sound/isa/opti9xx/opti92x-ad1848.c: In function 'snd_opti9xx_configure':sound/isa/opti9xx/opti92x-ad1848.c:322:43: error: overflow in conversion from 'int' to 'unsigned char' changes value from '(int)snd_opti9xx_read(chip, 3) & -256 | 240' to '240' [-Werror=overflow]
322 | (snd_opti9xx_read(chip, reg) & ~(mask)) | ((value) & (mask)))
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~
sound/isa/opti9xx/opti92x-ad1848.c:351:3: note: in expansion of macro 'snd_opti9xx_write_mask'
351 | snd_opti9xx_write_mask(chip, OPTi9XX_MC_REG(3), 0xf0, 0xff);
| ^~~~~~~~~~~~~~~~~~~~~~
sound/isa/opti9xx/miro.c: In function 'snd_miro_configure':
sound/isa/opti9xx/miro.c:873:40: error: overflow in conversion from 'int' to 'unsigned char' changes value from '(int)snd_miro_read(chip, 3) & -256 | 240' to '240' [-Werror=overflow]
873 | (snd_miro_read(chip, reg) & ~(mask)) | ((value) & (mask)))
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~
sound/isa/opti9xx/miro.c:1010:3: note: in expansion of macro 'snd_miro_write_mask'
1010 | snd_miro_write_mask(chip, OPTi9XX_MC_REG(3), 0xf0, 0xff);
| ^~~~~~~~~~~~~~~~~~~These are all harmless here as only the low 8 bit are passed down anyway. Change the macros to inline functions to make the code more readable and also avoid the warning.
Strictly speaking those functions also need locking to make the read/write pair atomic, but it seems unlikely that anyone would still run into that issue.
--- sound/isa/opti9xx/miro.c | 9 ++++++--- sound/isa/opti9xx/opti92x-ad1848.c | 9 ++++++--- 2 files changed, 12 insertions(+), 6 deletions(-)
sound/isa/opti9xx/miro.c +6 -3
@@ -867,10 +867,13 @@ static void snd_miro_write(struct snd_miro *chip, unsigned char reg,spin_unlock_irqrestore (&chip->lock, flags); } +static inline void snd_miro_write_mask (struct snd_miro *chip, +unsigned char reg,unsigned char value,unsigned char mask) +{ +unsigned char oldval =snd_miro_read (chip, reg); -#define snd_miro_write_mask (chip, reg, value, mask) \ - snd_miro_write(chip, reg, \ - (snd_miro_read(chip, reg) & ~(mask)) | ((value) & (mask))) +snd_miro_write (chip, reg, (oldval & ~mask) | (value & mask)); +}/* * Proc Interface
sound/isa/opti9xx/opti92x-ad1848.c +6 -3
@@ -317,10 +317,13 @@ static void snd_opti9xx_write(struct snd_opti9xx *chip, unsigned char reg, } -#define snd_opti9xx_write_mask (chip, reg, value, mask) \ - snd_opti9xx_write(chip, reg, \ - (snd_opti9xx_read(chip, reg) & ~(mask)) | ((value) & (mask))) +static inline void snd_opti9xx_write_mask (struct snd_opti9xx *chip, +unsigned char reg,unsigned char value,unsigned char mask) +{ +unsigned char oldval =snd_opti9xx_read (chip, reg); +snd_opti9xx_write (chip, reg, (oldval & ~mask) | (value & mask)); +} static int snd_opti9xx_configure(struct snd_opti9xx *chip, long port,
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 32/35 i2c: aspeed: Avoid i2c interrupt status clear race condition.
In reply to Sasha LevinFrom: ryan_chen <[email protected]>
[ Upstream commit c926c87b8e36dcc0ea5c2a0a0227ed4f32d0516a ]
In AST2600 there have a slow peripheral bus between CPU and i2c controller. Therefore GIC i2c interrupt status clear have delay timing, when CPU issue write clear i2c controller interrupt status. To avoid this issue, the driver need have read after write clear at i2c ISR.
--- drivers/i2c/busses/i2c-aspeed.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/i2c/busses/i2c-aspeed.c +4 -1
@@ -603,6 +603,7 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id)/* Ack all interrupts except for Rx done */ writel (irq_received & ~ASPEED_I2CD_INTR_RX_DONE , bus->base +ASPEED_I2C_INTR_STS_REG ); +readl (bus->base +ASPEED_I2C_INTR_STS_REG ); irq_remaining = irq_received;#if IS_ENABLED (CONFIG_I2C_SLAVE )
@@ -645,9 +646,11 @@ static irqreturn_t aspeed_i2c_bus_irq(int irq, void *dev_id) irq_received, irq_handled);/* Ack Rx done */ -if (irq_received &ASPEED_I2CD_INTR_RX_DONE ) +if (irq_received &ASPEED_I2CD_INTR_RX_DONE ) {writel (ASPEED_I2CD_INTR_RX_DONE , bus->base +ASPEED_I2C_INTR_STS_REG ); +readl (bus->base +ASPEED_I2C_INTR_STS_REG ); + }spin_unlock (&bus->lock);return irq_remaining ?IRQ_NONE :IRQ_HANDLED ; }
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 33/35 arm64: vdso: Add -fasynchronous-unwind-tables to cflags
In reply to Sasha LevinFrom: Vincenzo Frascino <[email protected]>
[ Upstream commit 1578e5d03112e3e9d37e1c4d95b6dfb734c73955 ]
On arm64 linux gcc uses -fasynchronous-unwind-tables -funwind-tables by default since gcc-8, so now the de facto platform ABI is to allow unwinding from async signal handlers.
However on bare metal targets (aarch64-none-elf), and on old gcc, async and sync unwind tables are not enabled by default to avoid runtime memory costs.
This means if linux is built with a baremetal toolchain the vdso.so may not have unwind tables which breaks the gcc platform ABI guarantee in userspace.
Add -fasynchronous-unwind-tables explicitly to the vgettimeofday.o cflags to address the ABI change.
--- arch/arm64/kernel/vdso/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/kernel/vdso/Makefile +1 -1
@@ -32,7 +32,7 @@ UBSAN_SANITIZE := n OBJECT_FILES_NON_STANDARD := y KCOV_INSTRUMENT := n -CFLAGS_vgettimeofday.o = -O2 -mcmodel=tiny +CFLAGS_vgettimeofday.o = -O2 -mcmodel=tiny -fasynchronous-unwind-tables ifneq ($(c-gettimeofday-y),) CFLAGS_vgettimeofday.o += -include $(c-gettimeofday-y)
Signature · 1 line
2.20.1
Sasha Levin
maintainerPATCH 34/35 iommu/amd: Fix legacy interrupt remapping for x2APIC-enabled system
In reply to Sasha LevinFrom: Suravee Suthikulpanit <[email protected]>
[ Upstream commit b74aa02d7a30ee5e262072a7d6e8deff10b37924 ]
Currently, system fails to boot because the legacy interrupt remapping mode does not enable 128-bit IRTE (GA), which is required for x2APIC support.
Fix by using AMD_IOMMU_GUEST_IR_LEGACY_GA mode when booting with kernel option amd_iommu_intr=legacy instead. The initialization logic will check GASup and automatically fallback to using AMD_IOMMU_GUEST_IR_LEGACY if GA mode is not supported.
--- drivers/iommu/amd_iommu_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/amd_iommu_init.c +1 -1
@@ -2946,7 +2946,7 @@ static int __init parse_amd_iommu_intr(char *str) {for (; *str; ++str) {if (strncmp (str,"legacy" ,6 ) ==0 ) { - amd_iommu_guest_ir =AMD_IOMMU_GUEST_IR_LEGACY ; + amd_iommu_guest_ir =AMD_IOMMU_GUEST_IR_LEGACY_GA ;break ; }if (strncmp (str,"vapic" ,5 ) ==0 ) {
Signature · 6 lines
2.20.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Sasha Levin
maintainerPATCH 35/35 iommu/qcom: Fix local_base status check
In reply to Sasha LevinFrom: Tang Bin <[email protected]>
[ Upstream commit b52649aee6243ea661905bdc5fbe28cc5f6dec76 ]
The function qcom_iommu_device_probe() does not perform sufficient error checking after executing devm_ioremap_resource(), which can result in crashes if a critical error path is encountered.
--- drivers/iommu/qcom_iommu.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/iommu/qcom_iommu.c +4 -1
@@ -814,8 +814,11 @@ static int qcom_iommu_device_probe(struct platform_device *pdev) qcom_iommu->dev = dev; res =platform_get_resource (pdev,IORESOURCE_MEM ,0 ); -if (res) +if (res) { qcom_iommu->local_base =devm_ioremap_resource (dev, res); +if (IS_ERR (qcom_iommu->local_base)) +return PTR_ERR (qcom_iommu->local_base); + } qcom_iommu->iface_clk =devm_clk_get (dev,"iface" );if (IS_ERR (qcom_iommu->iface_clk)) {
Signature · 6 lines
2.20.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu