- Kernel - mailweb.openeuler.org

[PATCH openEuler-1.0-LTS 1/5] rt2800: do not nullify initialization vector data
by Pu Lehui 18 Mar '24

18 Mar '24

From: Stanislaw Gruszka <sgruszka(a)redhat.com> mainline inclusion from mainline-v5.3-rc1 commit 710e6cc1595e25378c4b9977f7a8b4ad4a72a109 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I96GO9 CVE: CVE-2023-52595 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- If we restart hw we should keep existing IV (initialization vector) otherwise HW encryption will be broken after restart. Also fix some coding style issues on the way. Signed-off-by: Stanislaw Gruszka <sgruszka(a)redhat.com> Signed-off-by: Kalle Valo <kvalo(a)codeaurora.org> Signed-off-by: Pu Lehui <pulehui(a)huawei.com> --- drivers/net/wireless/ralink/rt2x00/rt2800lib.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c index a567bc273ffc..e499789674ca 100644 --- a/drivers/net/wireless/ralink/rt2x00/rt2800lib.c +++ b/drivers/net/wireless/ralink/rt2x00/rt2800lib.c @@ -1440,14 +1440,15 @@ static void rt2800_config_wcid_attr_cipher(struct rt2x00_dev *rt2x00dev, offset = MAC_IVEIV_ENTRY(key->hw_key_idx); - memset(&iveiv_entry, 0, sizeof(iveiv_entry)); + rt2800_register_multiread(rt2x00dev, offset, + &iveiv_entry, sizeof(iveiv_entry)); if ((crypto->cipher == CIPHER_TKIP) || (crypto->cipher == CIPHER_TKIP_NO_MIC) || (crypto->cipher == CIPHER_AES)) iveiv_entry.iv[3] |= 0x20; iveiv_entry.iv[3] |= key->keyidx << 6; rt2800_register_multiwrite(rt2x00dev, offset, - &iveiv_entry, sizeof(iveiv_entry)); + &iveiv_entry, sizeof(iveiv_entry)); } int rt2800_config_shared_key(struct rt2x00_dev *rt2x00dev, @@ -5546,13 +5547,11 @@ static int rt2800_init_registers(struct rt2x00_dev *rt2x00dev) * ASIC will keep garbage value after boot, clear encryption keys. */ for (i = 0; i < 4; i++) - rt2800_register_write(rt2x00dev, - SHARED_KEY_MODE_ENTRY(i), 0); + rt2800_register_write(rt2x00dev, SHARED_KEY_MODE_ENTRY(i), 0); for (i = 0; i < 256; i++) { rt2800_config_wcid(rt2x00dev, NULL, i); rt2800_delete_wcid_attr(rt2x00dev, i); - rt2800_register_write(rt2x00dev, MAC_IVEIV_ENTRY(i), 0); } /* -- 2.34.1

2 5

[PATCH openEuler-1.0-LTS] nfsd: fix use-after-free due to delegation race
by Li Lingfeng 18 Mar '24

18 Mar '24

From: "J. Bruce Fields" <bfields(a)redhat.com> mainline inclusion from mainline-v5.16-rc5 commit 548ec0805c399c65ed66c6641be467f717833ab5 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I99EU2 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- A delegation break could arrive as soon as we've called vfs_setlease. A delegation break runs a callback which immediately (in nfsd4_cb_recall_prepare) adds the delegation to del_recall_lru. If we then exit nfs4_set_delegation without hashing the delegation, it will be freed as soon as the callback is done with it, without ever being removed from del_recall_lru. Symptoms show up later as use-after-free or list corruption warnings, usually in the laundromat thread. I suspect aba2072f4523 "nfsd: grant read delegations to clients holding writes" made this bug easier to hit, but I looked as far back as v3.0 and it looks to me it already had the same problem. So I'm not sure where the bug was introduced; it may have been there from the beginning. Cc: stable(a)vger.kernel.org Signed-off-by: J. Bruce Fields <bfields(a)redhat.com> Conflict: fs/nfsd/nfs4state.c Commit 20b7d86f29d3("nfsd: use boottime for lease expiry calculation") get cl_time by ktime_get_boottime_seconds() instead of get_seconds(). Signed-off-by: Li Lingfeng <lilingfeng3(a)huawei.com> --- fs/nfsd/nfs4state.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index c8ce128e0054..d5d40aefd0a6 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -971,6 +971,11 @@ hash_delegation_locked(struct nfs4_delegation *dp, struct nfs4_file *fp) return 0; } +static bool delegation_hashed(struct nfs4_delegation *dp) +{ + return !(list_empty(&dp->dl_perfile)); +} + static bool unhash_delegation_locked(struct nfs4_delegation *dp) { @@ -978,7 +983,7 @@ unhash_delegation_locked(struct nfs4_delegation *dp) lockdep_assert_held(&state_lock); - if (list_empty(&dp->dl_perfile)) + if (!delegation_hashed(dp)) return false; dp->dl_stid.sc_type = NFS4_CLOSED_DELEG_STID; @@ -3901,7 +3906,7 @@ static void nfsd4_cb_recall_prepare(struct nfsd4_callback *cb) * queued for a lease break. Don't queue it again. */ spin_lock(&state_lock); - if (dp->dl_time == 0) { + if (delegation_hashed(dp) && dp->dl_time == 0) { dp->dl_time = get_seconds(); list_add_tail(&dp->dl_recall_lru, &nn->del_recall_lru); } -- 2.31.1

2 1

[PATCH openEuler-1.0-LTS 0/3] crypto: algif_aead - Only wake up when ctx->more is zero
by GUO Zihua 18 Mar '24

18 Mar '24

This patchset backports commit f3c802a1f300 ("crypto: algif_aead - Only wake up when ctx->more is zero") and following bugfixes. commit 21dfbcd1f5cb ("crypto: algif_aead - fix uninitialized ctx->init") is merged beforehand. Herbert Xu (3): crypto: algif_aead - Only wake up when ctx->more is zero crypto: af_alg - Fix regression on empty requests crypto: af_alg - Work around empty control messages without MSG_MORE crypto/af_alg.c | 22 +++++++++++++++++----- crypto/algif_aead.c | 4 ++-- crypto/algif_skcipher.c | 4 ++-- include/crypto/if_alg.h | 4 +++- 4 files changed, 24 insertions(+), 10 deletions(-) -- 2.34.1

2 4

[openeuler:OLK-6.6 1626/4034] drivers/char/ipmi/ipmi_si_ls2k500.c:92:5: sparse: sparse: symbol 'ipmi_si_sim_setup' was not declared. Should it be static?
by kernel test robot 18 Mar '24

18 Mar '24

tree: https://gitee.com/openeuler/kernel.git OLK-6.6 head: 23534451f50070452c470859343af52684963e9d commit: 4fbeaa84ef36adce930088074f8c088e5dd780c0 [1626/4034] ipmi: add ls2k500 bmc ipmi support. config: loongarch-randconfig-r123-20240317 (https://download.01.org/0day-ci/archive/20240318/202403181245.X9tjCWvK-lkp@…) compiler: loongarch64-linux-gcc (GCC) 13.2.0 reproduce: (https://download.01.org/0day-ci/archive/20240318/202403181245.X9tjCWvK-lkp@…) If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot <lkp(a)intel.com> | Closes: https://lore.kernel.org/oe-kbuild-all/202403181245.X9tjCWvK-lkp@intel.com/ sparse warnings: (new ones prefixed by >>) >> drivers/char/ipmi/ipmi_si_ls2k500.c:92:5: sparse: sparse: symbol 'ipmi_si_sim_setup' was not declared. Should it be static? >> drivers/char/ipmi/ipmi_si_ls2k500.c:116:29: sparse: sparse: incorrect type in assignment (different address spaces) @@ expected void *[addressable] [assigned] addr_source_data @@ got void [noderef] __iomem * @@ drivers/char/ipmi/ipmi_si_ls2k500.c:116:29: sparse: expected void *[addressable] [assigned] addr_source_data drivers/char/ipmi/ipmi_si_ls2k500.c:116:29: sparse: got void [noderef] __iomem * >> drivers/char/ipmi/ipmi_si_ls2k500.c:148:24: sparse: sparse: symbol 'ipmi_ls2k500_platform_driver' was not declared. Should it be static? drivers/char/ipmi/ipmi_si_ls2k500.c: note: in included file (through include/linux/mmzone.h, include/linux/gfp.h, include/linux/umh.h, include/linux/kmod.h, ...): include/linux/page-flags.h:242:46: sparse: sparse: self-comparison always evaluates to false vim +/ipmi_si_sim_setup +92 drivers/char/ipmi/ipmi_si_ls2k500.c 91 > 92 int ipmi_si_sim_setup(struct si_sm_io *io) 93 { 94 io->inputb = intf_sim_inb; 95 io->outputb = intf_sim_outb; 96 io->io_cleanup = ipmi_ls2k500_cleanup; 97 return 0; 98 } 99 100 #define platform_resource_start(dev, bar) ((dev)->resource[(bar)].start) 101 #define platform_resource_end(dev, bar) ((dev)->resource[(bar)].end) 102 static int of_ipmi_ls2k500_probe(struct platform_device *pdev) 103 { 104 int rv; 105 struct si_sm_io io; 106 void **kcs_data; 107 108 memset(&io, 0, sizeof(io)); 109 io.addr_source = SI_PLATFORM; 110 dev_info(&pdev->dev, "probing via ls2k500 platform"); 111 io.si_type = SI_KCS; 112 113 io.addr_space = IPMI_MEM_ADDR_SPACE; 114 io.io_setup = ipmi_si_sim_setup; 115 io.addr_data = pdev->resource[0].start; > 116 io.addr_source_data = ioremap(pdev->resource[0].start, 117 pdev->resource[0].end - 118 pdev->resource[0].start + 1); 119 kcs_data = dev_get_platdata(&pdev->dev); 120 event_jiffies = kcs_data[0]; 121 mscycles = kcs_data[1]; 122 io.dev = &pdev->dev; 123 io.regspacing = 4; 124 io.regsize = DEFAULT_REGSIZE; 125 io.regshift = 0; 126 io.irq = 0; 127 if (io.irq) 128 io.irq_setup = ipmi_std_irq_setup; 129 130 dev_info(&pdev->dev, "%pR regsize %d spacing %d irq %d\n", 131 &pdev->resource[0], io.regsize, io.regspacing, io.irq); 132 133 rv = ipmi_si_add_smi(&io); 134 if (rv) 135 ipmi_si_remove_by_dev(&pdev->dev); 136 137 return rv; 138 } 139 140 static int ipmi_ls2k500_remove(struct platform_device *pdev) 141 { 142 ipmi_si_remove_by_dev(&pdev->dev); 143 144 return 0; 145 } 146 147 #define LS2K500_SI_DEVICE_NAME "ipmi_ls2k500_si" > 148 struct platform_driver ipmi_ls2k500_platform_driver = { 149 .driver = { 150 .name = LS2K500_SI_DEVICE_NAME, 151 }, 152 .probe = of_ipmi_ls2k500_probe, 153 .remove = ipmi_ls2k500_remove, 154 }; 155 -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki

1 0

[PATCH OLK-5.10] serial: sc16is7xx: convert from _raw_ to _noinc_ regmap functions for FIFO
by GONG, Ruiqi 18 Mar '24

18 Mar '24

From: Hugo Villeneuve <hvilleneuve(a)dimonoff.com> stable inclusion from stable-v6.1.76 commit 416b10d2817c94db86829fb92ad43ce7d002c573 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I96ZNC CVE: CVE-2023-52488 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit dbf4ab821804df071c8b566d9813083125e6d97b upstream. The SC16IS7XX IC supports a burst mode to access the FIFOs where the initial register address is sent ($00), followed by all the FIFO data without having to resend the register address each time. In this mode, the IC doesn't increment the register address for each R/W byte. The regmap_raw_read() and regmap_raw_write() are functions which can perform IO over multiple registers. They are currently used to read/write from/to the FIFO, and although they operate correctly in this burst mode on the SPI bus, they would corrupt the regmap cache if it was not disabled manually. The reason is that when the R/W size is more than 1 byte, these functions assume that the register address is incremented and handle the cache accordingly. Convert FIFO R/W functions to use the regmap _noinc_ versions in order to remove the manual cache control which was a workaround when using the _raw_ versions. FIFO registers are properly declared as volatile so cache will not be used/updated for FIFO accesses. Fixes: dfeae619d781 ("serial: sc16is7xx") Cc: <stable(a)vger.kernel.org> Signed-off-by: Hugo Villeneuve <hvilleneuve(a)dimonoff.com> Link: https://lore.kernel.org/r/20231211171353.2901416-6-hugo@hugovil.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: drivers/tty/serial/sc16is7xx.c Signed-off-by: GONG, Ruiqi <gongruiqi1(a)huawei.com> --- drivers/tty/serial/sc16is7xx.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/drivers/tty/serial/sc16is7xx.c b/drivers/tty/serial/sc16is7xx.c index c681ed589303..5a08fe0df132 100644 --- a/drivers/tty/serial/sc16is7xx.c +++ b/drivers/tty/serial/sc16is7xx.c @@ -375,9 +375,7 @@ static void sc16is7xx_fifo_read(struct uart_port *port, unsigned int rxlen) const u8 line = sc16is7xx_line(port); u8 addr = (SC16IS7XX_RHR_REG << SC16IS7XX_REG_SHIFT) | line; - regcache_cache_bypass(s->regmap, true); - regmap_raw_read(s->regmap, addr, s->buf, rxlen); - regcache_cache_bypass(s->regmap, false); + regmap_noinc_read(s->regmap, addr, s->buf, rxlen); } static void sc16is7xx_fifo_write(struct uart_port *port, u8 to_send) @@ -393,9 +391,7 @@ static void sc16is7xx_fifo_write(struct uart_port *port, u8 to_send) if (unlikely(!to_send)) return; - regcache_cache_bypass(s->regmap, true); - regmap_raw_write(s->regmap, addr, s->buf, to_send); - regcache_cache_bypass(s->regmap, false); + regmap_noinc_write(s->regmap, addr, s->buf, to_send); } static void sc16is7xx_port_update(struct uart_port *port, u8 reg, @@ -488,6 +484,11 @@ static bool sc16is7xx_regmap_precious(struct device *dev, unsigned int reg) return false; } +static bool sc16is7xx_regmap_noinc(struct device *dev, unsigned int reg) +{ + return reg == SC16IS7XX_RHR_REG; +} + static int sc16is7xx_set_baud(struct uart_port *port, int baud) { struct sc16is7xx_port *s = dev_get_drvdata(port->dev); @@ -1426,6 +1427,8 @@ static struct regmap_config regcfg = { .cache_type = REGCACHE_RBTREE, .volatile_reg = sc16is7xx_regmap_volatile, .precious_reg = sc16is7xx_regmap_precious, + .writeable_noinc_reg = sc16is7xx_regmap_noinc, + .readable_noinc_reg = sc16is7xx_regmap_noinc, }; #ifdef CONFIG_SERIAL_SC16IS7XX_SPI -- 2.25.1

2 1

[PATCH OLK-6.6 v2 0/2] Fix CVE-2024-26629
by Li Lingfeng 18 Mar '24

18 Mar '24

The first patch is the fix patch of CVE. The second patch is the fix patch of the first one. NeilBrown (2): nfsd: fix RELEASE_LOCKOWNER nfsd: don't take fi_lock in nfsd_break_deleg_cb() fs/nfsd/nfs4state.c | 37 ++++++++++++++++++++----------------- 1 file changed, 20 insertions(+), 17 deletions(-) -- 2.31.1

2 3

[PATCH OLK-6.6 0/2] Fix CVE-2024-26629
by Li Lingfeng 18 Mar '24

18 Mar '24

The first patch is the fix patch of CVE. The second patch is the fix patch of the first one. NeilBrown (2): [Backport] nfsd: fix RELEASE_LOCKOWNER [Backport] nfsd: don't take fi_lock in nfsd_break_deleg_cb() fs/nfsd/nfs4state.c | 37 ++++++++++++++++++++----------------- 1 file changed, 20 insertions(+), 17 deletions(-) -- 2.31.1

2 3

[PATCH OLK-5.10 v2 0/4] Fix CVE-2024-26629
by Li Lingfeng 18 Mar '24

18 Mar '24

The first two patches are applied to reduce conflict. The third patch is the fix patch of CVE. The fourth patch is the fix patch of the third one. Chuck Lever (2): NFSD: Modernize nfsd4_release_lockowner() NFSD: Add documenting comment for nfsd4_release_lockowner() NeilBrown (2): nfsd: fix RELEASE_LOCKOWNER nfsd: don't take fi_lock in nfsd_break_deleg_cb() fs/nfsd/nfs4state.c | 76 ++++++++++++++++++++++++--------------------- 1 file changed, 41 insertions(+), 35 deletions(-) -- 2.31.1

2 5

[PATCH v3 OLK-6.6] iommu/iova: avoid softlockup in fq_flush_timeout
by Zhang Zekun 18 Mar '24

18 Mar '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8ZE0I CVE: NA -------------------------------- There is a softlockup undering cpu pressure test. ... pc : _raw_spin_unlock_irqrestore+0x14/0x78 lr : fq_flush_timeout+0x94/0x118 ... Call trace: _raw_spin_unlock_irqrestore+0x14/0x78 call_timer_fn+0x3c/0x1d0 expire_timers+0xcc/0x190 run_timer_softirq+0xfc/0x268 __do_softirq+0x128/0x3dc ____do_softirq+0x18/0x30 call_on_irq_stack+0x24/0x30 do_softirq_own_stack+0x24/0x38 irq_exit_rcu+0xc0/0xe8 el1_interrupt+0x48/0xc0 el1h_64_irq_handler+0x18/0x28 el1h_64_irq+0x78/0x80 __schedule+0xf28/0x12a0 schedule+0x3c/0x108 schedule_timeout+0xa0/0x1d0 pktgen_thread_worker+0x1180/0x15d0 kthread+0x120/0x130 ret_from_fork+0x10/0x20 This is because the timer callback fq_flush_timeout may run more than 10ms, and timer may be processed continuously in the softirq so trigger softlockup. We can use work to deal with fq_ring_free for each cpu which may take long time, that to avoid triggering softlockup. Signed-off-by: Li Bin <huawei.libin(a)huawei.com> Signed-off-by: Zhang Zekun <zhangzekun11(a)huawei.com> --- drivers/iommu/dma-iommu.c | 26 +++++++++++++++++++------- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c index 037fcf826407..58a90521630e 100644 --- a/drivers/iommu/dma-iommu.c +++ b/drivers/iommu/dma-iommu.c @@ -85,6 +85,8 @@ struct iommu_dma_cookie { /* Options for dma-iommu use */ struct iommu_dma_options options; struct mutex mutex; + + struct work_struct free_iova_work; }; static DEFINE_STATIC_KEY_FALSE(iommu_deferred_attach_enabled); @@ -184,17 +186,11 @@ static void fq_flush_iotlb(struct iommu_dma_cookie *cookie) static void fq_flush_timeout(struct timer_list *t) { struct iommu_dma_cookie *cookie = from_timer(cookie, t, fq_timer); - int cpu; atomic_set(&cookie->fq_timer_on, 0); fq_flush_iotlb(cookie); - if (cookie->options.qt == IOMMU_DMA_OPTS_SINGLE_QUEUE) { - fq_ring_free(cookie, cookie->single_fq); - } else { - for_each_possible_cpu(cpu) - fq_ring_free(cookie, per_cpu_ptr(cookie->percpu_fq, cpu)); - } + schedule_work(&cookie->free_iova_work); } static void queue_iova(struct iommu_dma_cookie *cookie, @@ -279,6 +275,7 @@ static void iommu_dma_free_fq(struct iommu_dma_cookie *cookie) return; del_timer_sync(&cookie->fq_timer); + flush_work(&cookie->free_iova_work); if (cookie->options.qt == IOMMU_DMA_OPTS_SINGLE_QUEUE) iommu_dma_free_fq_single(cookie->single_fq); else @@ -330,6 +327,20 @@ static int iommu_dma_init_fq_percpu(struct iommu_dma_cookie *cookie) return 0; } +static void free_iova_work_func(struct work_struct *work) +{ + struct iommu_dma_cookie *cookie = container_of(work, struct iommu_dma_cookie, free_iova_work); + int cpu; + + if (cookie->options.qt == IOMMU_DMA_OPTS_SINGLE_QUEUE) { + fq_ring_free(cookie, cookie->single_fq); + } else { + for_each_possible_cpu(cpu) + fq_ring_free(cookie, per_cpu_ptr(cookie->percpu_fq, cpu)); + } + +} + /* sysfs updates are serialised by the mutex of the group owning @domain */ int iommu_dma_init_fq(struct iommu_domain *domain) { @@ -352,6 +363,7 @@ int iommu_dma_init_fq(struct iommu_domain *domain) return -ENOMEM; } + INIT_WORK(&cookie->free_iova_work, free_iova_work_func); timer_setup(&cookie->fq_timer, fq_flush_timeout, 0); atomic_set(&cookie->fq_timer_on, 0); /* -- 2.17.1

2 1

[PATCH openEuler-1.0-LTS] printk: avoid deadlock in panic
by Ye Weihua 18 Mar '24

18 Mar '24

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I991OV -------------------------------- In the process of load testing against panic, a deadlock issue was discovered. The reason is that in the panic process, console_unlock doesn't handle the printk_context variable evenly, causing printk_context to overflow downwards, allowing vprintk_func to enter the nmi direct branch without disabling interrupt. Upon holding the logbuf_lock lock, an interrupt running wake_up_klogd_work_func was called, resulting in a deadlock. To avoid this issue, move printk_safe_ext_irqrestore after check abandon_console_lock_in_panic(). Fixes: 06a14c5f9e7d ("printk: Drop console_sem during panic") Signed-off-by: Ye Weihua <yeweihua4(a)huawei.com> --- kernel/printk/printk.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index f52eceb3c48a..11ebe66c473a 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -2426,7 +2426,7 @@ void console_unlock(void) static char text[LOG_LINE_MAX + PREFIX_MAX]; static int panic_console_dropped; unsigned long flags; - bool do_cond_resched, retry; + bool do_cond_resched, retry, locked = false; if (console_suspended) { up_console_sem(); @@ -2469,6 +2469,7 @@ void console_unlock(void) printk_safe_enter_irqsave(flags); raw_spin_lock(&logbuf_lock); + locked = true; if (console_seq < log_first_seq) { len = snprintf(text, sizeof(text), "** %llu printk messages dropped **\n", @@ -2520,6 +2521,7 @@ void console_unlock(void) } console_idx = log_next(console_idx); console_seq++; + locked = false; raw_spin_unlock(&logbuf_lock); /* @@ -2539,19 +2541,20 @@ void console_unlock(void) return; } - printk_safe_exit_irqrestore(flags); - /* Allow panic_cpu to take over the consoles safely */ if (abandon_console_lock_in_panic()) break; + printk_safe_exit_irqrestore(flags); + if (do_cond_resched) cond_resched(); } console_locked = 0; - raw_spin_unlock(&logbuf_lock); + if (likely(locked)) + raw_spin_unlock(&logbuf_lock); up_console_sem(); -- 2.34.1

2 1