- Linux-stable-mirror - lists.linaro.org

[PATCH v2 1/2] block: Introduce alloc_disk_node_attr()

by Bart Van Assche

This patch does not change the behavior of any existing code but introduces a function that will be used by the next patch. Signed-off-by: Bart Van Assche <bart.vanassche(a)wdc.com> Cc: Keith Busch <keith.busch(a)intel.com> Cc: Christoph Hellwig <hch(a)lst.de> Cc: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Cc: Sagi Grimberg <sagi(a)grimberg.me> Cc: Matias Bjorling <mb(a)lightnvm.io> Cc: <stable(a)vger.kernel.org> --- block/genhd.c | 26 ++++++++++++++++++++++++-- include/linux/genhd.h | 13 +++++++++++-- 2 files changed, 35 insertions(+), 4 deletions(-) diff --git a/block/genhd.c b/block/genhd.c index 8cc719a37b32..3b643152907d 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1413,10 +1413,20 @@ dev_t blk_lookup_devt(const char *name, int partno) } EXPORT_SYMBOL(blk_lookup_devt); -struct gendisk *__alloc_disk_node(int minors, int node_id) +/** + * __alloc_disk_node - allocate a gendisk structure and initialize it + * @minors: number of partitions to allocate memory for + * @node_id: NUMA node to allocate memory from + * @ag: NULL or pointer to a NULL-terminated array with attribute groups that + * must be attached to the disk_to_dev() of the allocated disk structure + */ +struct gendisk *__alloc_disk_node(int minors, int node_id, + const struct attribute_group **ag) { struct gendisk *disk; struct disk_part_tbl *ptbl; + const struct attribute_group **agp; + int num_groups = 0; if (minors > DISK_MAX_PARTS) { printk(KERN_ERR @@ -1425,7 +1435,15 @@ struct gendisk *__alloc_disk_node(int minors, int node_id) minors = DISK_MAX_PARTS; } - disk = kzalloc_node(sizeof(struct gendisk), GFP_KERNEL, node_id); + if (ag) { + for (agp = ag; *agp; agp++) + num_groups++; + /* Only count the terminating NULL if the array is not empty. */ + if (num_groups) + num_groups++; + } + disk = kzalloc_node(struct_size(disk, ag, num_groups), GFP_KERNEL, + node_id); if (disk) { if (!init_part_stats(&disk->part0)) { kfree(disk); @@ -1461,6 +1479,10 @@ struct gendisk *__alloc_disk_node(int minors, int node_id) rand_initialize_disk(disk); disk_to_dev(disk)->class = &block_class; disk_to_dev(disk)->type = &disk_type; + if (num_groups) { + memcpy(disk->ag, ag, num_groups * sizeof(ag)); + disk_to_dev(disk)->groups = disk->ag; + } device_initialize(disk_to_dev(disk)); } return disk; diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 57864422a2c8..50ab6cf77fdf 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -211,6 +211,11 @@ struct gendisk { int node_id; struct badblocks *bb; struct lockdep_map lockdep_map; + /* + * Storage space for NULL-terminated attribute group array for + * disk_to_dev() of this gendisk structure. + */ + const struct attribute_group *ag[0]; }; static inline struct gendisk *part_to_disk(struct hd_struct *part) @@ -608,7 +613,8 @@ extern void __delete_partition(struct percpu_ref *); extern void delete_partition(struct gendisk *, int); extern void printk_all_partitions(void); -extern struct gendisk *__alloc_disk_node(int minors, int node_id); +extern struct gendisk *__alloc_disk_node(int minors, int node_id, + const struct attribute_group **ag); extern struct kobject *get_disk_and_module(struct gendisk *disk); extern void put_disk(struct gendisk *disk); extern void put_disk_and_module(struct gendisk *disk); @@ -634,6 +640,9 @@ extern ssize_t part_fail_store(struct device *dev, #endif /* CONFIG_FAIL_MAKE_REQUEST */ #define alloc_disk_node(minors, node_id) \ + alloc_disk_node_attr(minors, node_id, NULL) + +#define alloc_disk_node_attr(minors, node_id, ag) \ ({ \ static struct lock_class_key __key; \ const char *__name; \ @@ -641,7 +650,7 @@ extern ssize_t part_fail_store(struct device *dev, \ __name = "(gendisk_completion)"#minors"("#node_id")"; \ \ - __disk = __alloc_disk_node(minors, node_id); \ + __disk = __alloc_disk_node(minors, node_id, ag); \ \ if (__disk) \ lockdep_init_map(&__disk->lockdep_map, __name, &__key, 0); \ -- 2.18.0

7 years, 4 months

2
1
0 0

[PATCH 0/3 stable 4.9] proc/sysctl: prune stale dentries during unregistering

by Mike Rapoport

Hi, These patches fix a memory leak in per-cpu areas in v4.9.y The following command sequence while :; do c=$(docker create busybox) docker start $c docker stop $c docker rm $c done causes the number of pcpu_get_vm_area entries in /proc/vmallocinfo to grow. The commit d6cffbbe9a7e ("proc/sysctl: prune stale dentries during unregistering") fixes the issue and the commits ace0c791e6c3 ("proc/sysctl: Don't grab i_lock under sysctl_lock.") and 2fd1d2c4ceb2 ("proc: Fix proc_sys_prune_dcache to hold a sb reference") are the follow-up fixups. I've also checked v4.4 and the issue does not appear there. Eric W. Biederman (2): proc/sysctl: Don't grab i_lock under sysctl_lock. proc: Fix proc_sys_prune_dcache to hold a sb reference Konstantin Khlebnikov (1): proc/sysctl: prune stale dentries during unregistering fs/proc/inode.c | 3 +- fs/proc/internal.h | 7 +++-- fs/proc/proc_sysctl.c | 83 +++++++++++++++++++++++++++++++++++++++----------- include/linux/sysctl.h | 1 + 4 files changed, 74 insertions(+), 20 deletions(-) -- 2.7.4

7 years, 4 months

2
8
0 0

[PATCH for-4.14.y 1/5] mtd: nand: qcom: Add a NULL check for devm_kasprintf()

by Amit Pundir

From: Fabio Estevam <fabio.estevam(a)nxp.com> commit 069f05346d01e7298939f16533953cdf52370be3 upstream. devm_kasprintf() may fail, so we should better add a NULL check and propagate an error on failure. Signed-off-by: Fabio Estevam <fabio.estevam(a)nxp.com> Signed-off-by: Boris Brezillon <boris.brezillon(a)free-electrons.com> Signed-off-by: Amit Pundir <amit.pundir(a)linaro.org> --- Cherry-picked from lede tree https://git.lede-project.org/?p=source.git Please apply it on 4.9.y as well. Not applicable for 4.4.y and 3.18.y drivers/mtd/nand/qcom_nandc.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/mtd/nand/qcom_nandc.c b/drivers/mtd/nand/qcom_nandc.c index 3baddfc997d1..b49ca02b399d 100644 --- a/drivers/mtd/nand/qcom_nandc.c +++ b/drivers/mtd/nand/qcom_nandc.c @@ -2544,6 +2544,9 @@ static int qcom_nand_host_init(struct qcom_nand_controller *nandc, nand_set_flash_node(chip, dn); mtd->name = devm_kasprintf(dev, GFP_KERNEL, "qcom_nand.%d", host->cs); + if (!mtd->name) + return -ENOMEM; + mtd->owner = THIS_MODULE; mtd->dev.parent = dev; -- 2.7.4

7 years, 4 months

2
6
0 0

request for 4.4-stable: d8f9cc328c88 ("IB/mlx4: Mark user MR as writable if actual virtual memory is writable")

by Sudip Mukherjee

Hi Greg, This was missing in 4.4-stable. Please apply to your queue. It will depend on 08bb558ac11a ("IB/core: Make testing MR flags for writability a static inline function") which I have sent earlier. -- Regards Sudip

7 years, 4 months

2
1
0 0

[PATCH 1/5] Tools: hv: Fix a bug in the key delete code

by kys＠linuxonhyperv.com

From: "K. Y. Srinivasan" <kys(a)microsoft.com> Fix a bug in the key delete code - the num_records range from 0 to num_records-1. Signed-off-by: K. Y. Srinivasan <kys(a)microsoft.com> Reported-by: David Binderman <dcb314(a)hotmail.com> Cc: <stable(a)vger.kernel.org> --- tools/hv/hv_kvp_daemon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/hv/hv_kvp_daemon.c b/tools/hv/hv_kvp_daemon.c index dbf6e8bd98ba..bbb2a8ef367c 100644 --- a/tools/hv/hv_kvp_daemon.c +++ b/tools/hv/hv_kvp_daemon.c @@ -286,7 +286,7 @@ static int kvp_key_delete(int pool, const __u8 *key, int key_size) * Found a match; just move the remaining * entries up. */ - if (i == num_records) { + if (i == (num_records - 1)) { kvp_file_info[pool].num_records--; kvp_update_file(pool); return 0; -- 2.18.0

7 years, 4 months

2
1
0 0

[Linux-stable-mirror] [PATCH] soc/tegra: pmc: fix child-node lookup

by Johan Hovold

Fix child-node lookup during probe, which ended up searching the whole device tree depth-first starting at the parent rather than just matching on its children. To make things worse, the parent pmc node could end up being prematurely freed as of_find_node_by_name() drops a reference to its first argument. Fixes: 3568df3d31d6 ("soc: tegra: Add thermal reset (thermtrip) support to PMC") Cc: stable <stable(a)vger.kernel.org> # 4.0 Cc: Mikko Perttunen <mperttunen(a)nvidia.com> Signed-off-by: Johan Hovold <johan(a)kernel.org> --- drivers/soc/tegra/pmc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soc/tegra/pmc.c b/drivers/soc/tegra/pmc.c index 0453ff6839a7..7e9ef3431bea 100644 --- a/drivers/soc/tegra/pmc.c +++ b/drivers/soc/tegra/pmc.c @@ -1321,7 +1321,7 @@ static void tegra_pmc_init_tsense_reset(struct tegra_pmc *pmc) if (!pmc->soc->has_tsense_reset) return; - np = of_find_node_by_name(pmc->dev->of_node, "i2c-thermtrip"); + np = of_get_child_by_name(pmc->dev->of_node, "i2c-thermtrip"); if (!np) { dev_warn(dev, "i2c-thermtrip node not found, %s.\n", disabled); return; -- 2.15.0

7 years, 4 months

3
5
0 0

[PATCH v7 0/9] powerpc/pseries: Machine check handler improvements.

by Mahesh J Salgaonkar

This patch series includes some improvement to Machine check handler for pseries. Patch 1 fixes a buffer overrun issue if rtas extended error log size is greater than RTAS_ERROR_LOG_MAX. Patch 2 fixes an issue where machine check handler crashes kernel while accessing vmalloc-ed buffer while in nmi context. Patch 3 fixes endain bug while restoring of r3 in MCE handler. Patch 5 implements a real mode mce handler and flushes the SLBs on SLB error. Patch 6 display's the MCE error details on console. Patch 7 saves and dumps the SLB contents on SLB MCE errors to improve the debugability. Patch 8 adds sysctl knob for recovery action on recovered MCEs. Patch 9 consolidates mce early real mode handling code. Change in V7: - Fold Michal's patch into patch 5 - Handle MSR_RI=0 and evil context case in MC handler in patch 5. - Patch 7: Print slb cache ptr value and slb cache data. - Move patch 8 to patch 9. - Introduce patch 8 add sysctl knob for recovery action on recovered MCEs. Change in V6: - Introduce patch 8 to consolidate early real mode handling code. - Address Nick's comment on erroneous hunk. Change in V5: - Use min_t instead of max_t. - Fix an issue reported by kbuild test robot and address review comments. Change in V4: - Flush the SLBs in real mode mce handler to handle SLB errors for entry 0. - Allocate buffers per cpu to hold rtas error log and old slb contents. - Defer the logging of rtas error log to irq work queue. Change in V3: - Moved patch 5 to patch 2 Change in V2: - patch 3: Display additional info (NIP and task info) in MCE error details. - patch 5: Fix endain bug while restoring of r3 in MCE handler. --- Mahesh Salgaonkar (9): powerpc/pseries: Avoid using the size greater than RTAS_ERROR_LOG_MAX. powerpc/pseries: Defer the logging of rtas error to irq work queue. powerpc/pseries: Fix endainness while restoring of r3 in MCE handler. powerpc/pseries: Define MCE error event section. powerpc/pseries: flush SLB contents on SLB MCE errors. powerpc/pseries: Display machine check error details. powerpc/pseries: Dump the SLB contents on SLB MCE errors. powerpc/mce: Add sysctl control for recovery action on MCE. powernv/pseries: consolidate code for mce early handling. arch/powerpc/include/asm/book3s/64/mmu-hash.h | 8 + arch/powerpc/include/asm/machdep.h | 1 arch/powerpc/include/asm/mce.h | 2 arch/powerpc/include/asm/paca.h | 7 + arch/powerpc/include/asm/rtas.h | 116 ++++++++++++ arch/powerpc/kernel/exceptions-64s.S | 42 ++++ arch/powerpc/kernel/mce.c | 73 +++++++- arch/powerpc/kernel/traps.c | 3 arch/powerpc/mm/slb.c | 79 ++++++++ arch/powerpc/platforms/powernv/setup.c | 15 ++ arch/powerpc/platforms/pseries/pseries.h | 1 arch/powerpc/platforms/pseries/ras.c | 242 +++++++++++++++++++++++-- arch/powerpc/platforms/pseries/setup.c | 27 +++ 13 files changed, 588 insertions(+), 28 deletions(-) -- Signature

7 years, 4 months

2
3
0 0

[PATCH v2 1/2] powerpc/fadump: handle crash memory ranges array index overflow

by Hari Bathini

Crash memory ranges is an array of memory ranges of the crashing kernel to be exported as a dump via /proc/vmcore file. The size of the array is set based on INIT_MEMBLOCK_REGIONS, which works alright in most cases where memblock memory regions count is less than INIT_MEMBLOCK_REGIONS value. But this count can grow beyond INIT_MEMBLOCK_REGIONS value since commit 142b45a72e22 ("memblock: Add array resizing support"). On large memory systems with a few DLPAR operations, the memblock memory regions count could be larger than INIT_MEMBLOCK_REGIONS value. On such systems, registering fadump results in crash or other system failures like below: task: c00007f39a290010 ti: c00000000b738000 task.ti: c00000000b738000 NIP: c000000000047df4 LR: c0000000000f9e58 CTR: c00000000010f180 REGS: c00000000b73b570 TRAP: 0300 Tainted: G L X (4.4.140+) MSR: 8000000000009033 <SF,EE,ME,IR,DR,RI,LE> CR: 22004484 XER: 20000000 CFAR: c000000000008500 DAR: 000007a450000000 DSISR: 40000000 SOFTE: 0 GPR00: c0000000000f9e58 c00000000b73b7f0 c000000000f09a00 000000000000001a GPR04: c00007f3bf774c90 0000000000000004 c000000000eb9a00 0000000000000800 GPR08: 0000000000000804 000007a450000000 c000000000fa9a00 c00007ffb169ca20 GPR12: 0000000022004482 c00000000fa12c00 c00007f3a0ea97a8 0000000000000000 GPR16: c00007f3a0ea9a50 c00000000b73bd60 0000000000000118 000000000001fe80 GPR20: 0000000000000118 0000000000000000 c000000000b8c980 00000000000000d0 GPR24: 000007ffb0b10000 c00007ffb169c980 0000000000000000 c000000000b8c980 GPR28: 0000000000000004 c00007ffb169c980 000000000000001a c00007ffb169c980 NIP [c000000000047df4] smp_send_reschedule+0x24/0x80 LR [c0000000000f9e58] resched_curr+0x138/0x160 Call Trace: [c00000000b73b7f0] [c0000000000f9e58] resched_curr+0x138/0x160 (unreliable) [c00000000b73b820] [c0000000000fb538] check_preempt_curr+0xc8/0xf0 [c00000000b73b850] [c0000000000fb598] ttwu_do_wakeup+0x38/0x150 [c00000000b73b890] [c0000000000fc9c4] try_to_wake_up+0x224/0x4d0 [c00000000b73b900] [c00000000011ef34] __wake_up_common+0x94/0x100 [c00000000b73b960] [c00000000034a78c] ep_poll_callback+0xac/0x1c0 [c00000000b73b9b0] [c00000000011ef34] __wake_up_common+0x94/0x100 [c00000000b73ba10] [c00000000011f810] __wake_up_sync_key+0x70/0xa0 [c00000000b73ba60] [c00000000067c3e8] sock_def_readable+0x58/0xa0 [c00000000b73ba90] [c0000000007848ac] unix_stream_sendmsg+0x2dc/0x4c0 [c00000000b73bb70] [c000000000675a38] sock_sendmsg+0x68/0xa0 [c00000000b73bba0] [c00000000067673c] ___sys_sendmsg+0x2cc/0x2e0 [c00000000b73bd30] [c000000000677dbc] __sys_sendmsg+0x5c/0xc0 [c00000000b73bdd0] [c0000000006789bc] SyS_socketcall+0x36c/0x3f0 [c00000000b73be30] [c000000000009488] system_call+0x3c/0x100 Instruction dump: 4e800020 60000000 60420000 3c4c00ec 38421c30 7c0802a6 f8010010 60000000 3d42000a e92ab420 2fa90000 4dde0020 <e9290000> 2fa90000 419e0044 7c0802a6 ---[ end trace a6d1dd4bab5f8253 ]--- as array index overflow is not checked for while setting up crash memory ranges causing memory corruption. To resolve this issue, dynamically allocate memory for crash memory ranges and resize it incrementally, in units of pagesize, on hitting array size limit. Fixes: 2df173d9e85d ("fadump: Initialize elfcore header and add PT_LOAD program headers.") Cc: stable(a)vger.kernel.org Cc: Mahesh Salgaonkar <mahesh(a)linux.vnet.ibm.com> Signed-off-by: Hari Bathini <hbathini(a)linux.ibm.com> --- Changes in v2: * Allocating memory for crash ranges in pagesize unit. * freeing memory allocated while cleaning up. * Moved the changes to coalesce memory ranges into patch 2/2. arch/powerpc/include/asm/fadump.h | 4 +- arch/powerpc/kernel/fadump.c | 91 +++++++++++++++++++++++++++++++------ 2 files changed, 79 insertions(+), 16 deletions(-) diff --git a/arch/powerpc/include/asm/fadump.h b/arch/powerpc/include/asm/fadump.h index 5a23010..3abc738 100644 --- a/arch/powerpc/include/asm/fadump.h +++ b/arch/powerpc/include/asm/fadump.h @@ -195,8 +195,8 @@ struct fadump_crash_info_header { struct cpumask online_mask; }; -/* Crash memory ranges */ -#define INIT_CRASHMEM_RANGES (INIT_MEMBLOCK_REGIONS + 2) +/* Crash memory ranges size unit (pagesize) */ +#define CRASHMEM_RANGES_ALLOC_SIZE PAGE_SIZE struct fad_crash_memory_ranges { unsigned long long base; diff --git a/arch/powerpc/kernel/fadump.c b/arch/powerpc/kernel/fadump.c index 07e8396..2ec5704 100644 --- a/arch/powerpc/kernel/fadump.c +++ b/arch/powerpc/kernel/fadump.c @@ -47,8 +47,10 @@ static struct fadump_mem_struct fdm; static const struct fadump_mem_struct *fdm_active; static DEFINE_MUTEX(fadump_mutex); -struct fad_crash_memory_ranges crash_memory_ranges[INIT_CRASHMEM_RANGES]; +struct fad_crash_memory_ranges *crash_memory_ranges; +int crash_memory_ranges_size; int crash_mem_ranges; +int max_crash_mem_ranges; /* Scan the Firmware Assisted dump configuration details. */ int __init early_init_dt_scan_fw_dump(unsigned long node, @@ -868,22 +870,67 @@ static int __init process_fadump(const struct fadump_mem_struct *fdm_active) return 0; } -static inline void fadump_add_crash_memory(unsigned long long base, - unsigned long long end) +static void free_crash_memory_ranges(void) +{ + kfree(crash_memory_ranges); + crash_memory_ranges = NULL; + crash_memory_ranges_size = 0; + max_crash_mem_ranges = 0; +} + +/* + * Allocate or reallocate crash memory ranges array in incremental units + * of CRASHMEM_RANGES_ALLOC_SIZE. + */ +static int allocate_crash_memory_ranges(void) +{ + u64 new_size; + struct fad_crash_memory_ranges *new_array; + + new_size = crash_memory_ranges_size + CRASHMEM_RANGES_ALLOC_SIZE; + pr_debug("Allocating %llu bytes of memory for crash memory ranges\n", + new_size); + + new_array = krealloc(crash_memory_ranges, new_size, GFP_KERNEL); + if (new_array == NULL) { + pr_err("Insufficient memory for setting up crash memory ranges\n"); + free_crash_memory_ranges(); + return -ENOMEM; + } + + crash_memory_ranges = new_array; + crash_memory_ranges_size = new_size; + max_crash_mem_ranges = (new_size / + sizeof(struct fad_crash_memory_ranges)); + return 0; +} + +static inline int fadump_add_crash_memory(unsigned long long base, + unsigned long long end) { if (base == end) - return; + return 0; + + if (crash_mem_ranges == max_crash_mem_ranges) { + int ret; + + ret = allocate_crash_memory_ranges(); + if (ret) + return ret; + } pr_debug("crash_memory_range[%d] [%#016llx-%#016llx], %#llx bytes\n", crash_mem_ranges, base, end - 1, (end - base)); crash_memory_ranges[crash_mem_ranges].base = base; crash_memory_ranges[crash_mem_ranges].size = end - base; crash_mem_ranges++; + return 0; } -static void fadump_exclude_reserved_area(unsigned long long start, +static int fadump_exclude_reserved_area(unsigned long long start, unsigned long long end) { + int ret = 0; unsigned long long ra_start, ra_end; ra_start = fw_dump.reserve_dump_area_start; @@ -891,15 +938,20 @@ static void fadump_exclude_reserved_area(unsigned long long start, if ((ra_start < end) && (ra_end > start)) { if ((start < ra_start) && (end > ra_end)) { - fadump_add_crash_memory(start, ra_start); - fadump_add_crash_memory(ra_end, end); + ret = fadump_add_crash_memory(start, ra_start); + if (ret) + return ret; + + ret = fadump_add_crash_memory(ra_end, end); } else if (start < ra_start) { - fadump_add_crash_memory(start, ra_start); + ret = fadump_add_crash_memory(start, ra_start); } else if (ra_end < end) { - fadump_add_crash_memory(ra_end, end); + ret = fadump_add_crash_memory(ra_end, end); } } else - fadump_add_crash_memory(start, end); + ret = fadump_add_crash_memory(start, end); + + return ret; } static int fadump_init_elfcore_header(char *bufp) @@ -939,8 +991,9 @@ static int fadump_init_elfcore_header(char *bufp) * Traverse through memblock structure and setup crash memory ranges. These * ranges will be used create PT_LOAD program headers in elfcore header. */ -static void fadump_setup_crash_memory_ranges(void) +static int fadump_setup_crash_memory_ranges(void) { + int ret; struct memblock_region *reg; unsigned long long start, end; @@ -953,7 +1006,9 @@ static void fadump_setup_crash_memory_ranges(void) * specified during fadump registration. We need to create a separate * program header for this chunk with the correct offset. */ - fadump_add_crash_memory(RMA_START, fw_dump.boot_memory_size); + ret = fadump_add_crash_memory(RMA_START, fw_dump.boot_memory_size); + if (ret) + return ret; for_each_memblock(memory, reg) { start = (unsigned long long)reg->base; @@ -973,8 +1028,12 @@ static void fadump_setup_crash_memory_ranges(void) } /* add this range excluding the reserved dump area. */ - fadump_exclude_reserved_area(start, end); + ret = fadump_exclude_reserved_area(start, end); + if (ret) + return ret; } + + return 0; } /* @@ -1095,6 +1154,7 @@ static unsigned long init_fadump_header(unsigned long addr) static int register_fadump(void) { + int ret; unsigned long addr; void *vaddr; @@ -1105,7 +1165,9 @@ static int register_fadump(void) if (!fw_dump.reserve_dump_area_size) return -ENODEV; - fadump_setup_crash_memory_ranges(); + ret = fadump_setup_crash_memory_ranges(); + if (ret) + return ret; addr = be64_to_cpu(fdm.rmr_region.destination_address) + be64_to_cpu(fdm.rmr_region.source_len); /* Initialize fadump crash info header. */ @@ -1183,6 +1245,7 @@ void fadump_cleanup(void) } else if (fw_dump.dump_registered) { /* Un-register Firmware-assisted dump if it was registered. */ fadump_unregister_dump(&fdm); + free_crash_memory_ranges(); } }

7 years, 4 months

3
2
0 0

[PATCH v3 2/2] KVM: arm/arm64: Skip updating PTE entry if no change

by Punit Agrawal

When there is contention on faulting in a particular page table entry at stage 2, the break-before-make requirement of the architecture can lead to additional refaulting due to TLB invalidation. Avoid this by skipping a page table update if the new value of the PTE matches the previous value. Fixes: d5d8184d35c9 ("KVM: ARM: Memory virtualization setup") Signed-off-by: Punit Agrawal <punit.agrawal(a)arm.com> Reviewed-by: Suzuki Poulose <suzuki.poulose(a)arm.com> Cc: Marc Zyngier <marc.zyngier(a)arm.com> Cc: Christoffer Dall <christoffer.dall(a)arm.com> Cc: stable(a)vger.kernel.org --- virt/kvm/arm/mmu.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c index 2bb0b5dba412..c2b95a22959b 100644 --- a/virt/kvm/arm/mmu.c +++ b/virt/kvm/arm/mmu.c @@ -1118,6 +1118,10 @@ static int stage2_set_pte(struct kvm *kvm, struct kvm_mmu_memory_cache *cache, /* Create 2nd stage page table mapping - Level 3 */ old_pte = *pte; if (pte_present(old_pte)) { + /* Skip page table update if there is no change */ + if (pte_val(old_pte) == pte_val(*new_pte)) + return 0; + kvm_set_pte(pte, __pte(0)); kvm_tlb_flush_vmid_ipa(kvm, addr); } else { -- 2.18.0

7 years, 4 months

1
0
0 0

[PATCH v3 1/2] KVM: arm/arm64: Skip updating PMD entry if no change

by Punit Agrawal

Contention on updating a PMD entry by a large number of vcpus can lead to duplicate work when handling stage 2 page faults. As the page table update follows the break-before-make requirement of the architecture, it can lead to repeated refaults due to clearing the entry and flushing the tlbs. This problem is more likely when - * there are large number of vcpus * the mapping is large block mapping such as when using PMD hugepages (512MB) with 64k pages. Fix this by skipping the page table update if there is no change in the entry being updated. Fixes: ad361f093c1e ("KVM: ARM: Support hugetlbfs backed huge pages") Signed-off-by: Punit Agrawal <punit.agrawal(a)arm.com> Reviewed-by: Suzuki Poulose <suzuki.poulose(a)arm.com> Cc: Marc Zyngier <marc.zyngier(a)arm.com> Cc: Christoffer Dall <christoffer.dall(a)arm.com> Cc: stable(a)vger.kernel.org --- virt/kvm/arm/mmu.c | 38 +++++++++++++++++++++++++++----------- 1 file changed, 27 insertions(+), 11 deletions(-) diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c index 1d90d79706bd..2bb0b5dba412 100644 --- a/virt/kvm/arm/mmu.c +++ b/virt/kvm/arm/mmu.c @@ -1015,19 +1015,35 @@ static int stage2_set_pmd_huge(struct kvm *kvm, struct kvm_mmu_memory_cache pmd = stage2_get_pmd(kvm, cache, addr); VM_BUG_ON(!pmd); - /* - * Mapping in huge pages should only happen through a fault. If a - * page is merged into a transparent huge page, the individual - * subpages of that huge page should be unmapped through MMU - * notifiers before we get here. - * - * Merging of CompoundPages is not supported; they should become - * splitting first, unmapped, merged, and mapped back in on-demand. - */ - VM_BUG_ON(pmd_present(*pmd) && pmd_pfn(*pmd) != pmd_pfn(*new_pmd)); - old_pmd = *pmd; if (pmd_present(old_pmd)) { + /* + * Multiple vcpus faulting on the same PMD entry, can + * lead to them sequentially updating the PMD with the + * same value. Following the break-before-make + * (pmd_clear() followed by tlb_flush()) process can + * hinder forward progress due to refaults generated + * on missing translations. + * + * Skip updating the page table if the entry is + * unchanged. + */ + if (pmd_val(old_pmd) == pmd_val(*new_pmd)) + return 0; + + /* + * Mapping in huge pages should only happen through a + * fault. If a page is merged into a transparent huge + * page, the individual subpages of that huge page + * should be unmapped through MMU notifiers before we + * get here. + * + * Merging of CompoundPages is not supported; they + * should become splitting first, unmapped, merged, + * and mapped back in on-demand. + */ + VM_BUG_ON(pmd_pfn(old_pmd) != pmd_pfn(*new_pmd)); + pmd_clear(pmd); kvm_tlb_flush_vmid_ipa(kvm, addr); } else { -- 2.18.0

7 years, 4 months

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror