November 2018 - Linux-stable-mirror

FAILED: patch "[PATCH] acpi/nfit, x86/mce: Handle only uncorrectable machine checks" failed to apply to 4.9-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.9-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 5d96c9342c23ee1d084802dcf064caa67ecaa45b Mon Sep 17 00:00:00 2001 From: Vishal Verma <vishal.l.verma(a)intel.com> Date: Thu, 25 Oct 2018 18:37:28 -0600 Subject: [PATCH] acpi/nfit, x86/mce: Handle only uncorrectable machine checks The MCE handler for nfit devices is called for memory errors on a Non-Volatile DIMM and adds the error location to a 'badblocks' list. This list is used by the various NVDIMM drivers to avoid consuming known poison locations during IO. The MCE handler gets called for both corrected and uncorrectable errors. Until now, both kinds of errors have been added to the badblocks list. However, corrected memory errors indicate that the problem has already been fixed by hardware, and the resulting interrupt is merely a notification to Linux. As far as future accesses to that location are concerned, it is perfectly fine to use, and thus doesn't need to be included in the above badblocks list. Add a check in the nfit MCE handler to filter out corrected mce events, and only process uncorrectable errors. Fixes: 6839a6d96f4e ("nfit: do an ARS scrub on hitting a latent media error") Reported-by: Omar Avelar <omar.avelar(a)intel.com> Signed-off-by: Vishal Verma <vishal.l.verma(a)intel.com> Signed-off-by: Borislav Petkov <bp(a)suse.de> CC: Arnd Bergmann <arnd(a)arndb.de> CC: Dan Williams <dan.j.williams(a)intel.com> CC: Dave Jiang <dave.jiang(a)intel.com> CC: elliott(a)hpe.com CC: "H. Peter Anvin" <hpa(a)zytor.com> CC: Ingo Molnar <mingo(a)redhat.com> CC: Len Brown <lenb(a)kernel.org> CC: linux-acpi(a)vger.kernel.org CC: linux-edac <linux-edac(a)vger.kernel.org> CC: linux-nvdimm(a)lists.01.org CC: Qiuxu Zhuo <qiuxu.zhuo(a)intel.com> CC: "Rafael J. Wysocki" <rjw(a)rjwysocki.net> CC: Ross Zwisler <zwisler(a)kernel.org> CC: stable <stable(a)vger.kernel.org> CC: Thomas Gleixner <tglx(a)linutronix.de> CC: Tony Luck <tony.luck(a)intel.com> CC: x86-ml <x86(a)kernel.org> CC: Yazen Ghannam <yazen.ghannam(a)amd.com> Link: http://lkml.kernel.org/r/20181026003729.8420-1-vishal.l.verma@intel.com diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 4da9b1c58d28..dbd9fe2f6163 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -221,6 +221,7 @@ static inline void mce_hygon_feature_init(struct cpuinfo_x86 *c) { return mce_am int mce_available(struct cpuinfo_x86 *c); bool mce_is_memory_error(struct mce *m); +bool mce_is_correctable(struct mce *m); DECLARE_PER_CPU(unsigned, mce_exception_count); DECLARE_PER_CPU(unsigned, mce_poll_count); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 8c66d2fc8f81..77527b8ea982 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -534,7 +534,7 @@ bool mce_is_memory_error(struct mce *m) } EXPORT_SYMBOL_GPL(mce_is_memory_error); -static bool mce_is_correctable(struct mce *m) +bool mce_is_correctable(struct mce *m) { if (m->cpuvendor == X86_VENDOR_AMD && m->status & MCI_STATUS_DEFERRED) return false; @@ -547,6 +547,7 @@ static bool mce_is_correctable(struct mce *m) return true; } +EXPORT_SYMBOL_GPL(mce_is_correctable); static bool cec_add_mce(struct mce *m) { diff --git a/drivers/acpi/nfit/mce.c b/drivers/acpi/nfit/mce.c index e9626bf6ca29..7a51707f87e9 100644 --- a/drivers/acpi/nfit/mce.c +++ b/drivers/acpi/nfit/mce.c @@ -25,8 +25,8 @@ static int nfit_handle_mce(struct notifier_block *nb, unsigned long val, struct acpi_nfit_desc *acpi_desc; struct nfit_spa *nfit_spa; - /* We only care about memory errors */ - if (!mce_is_memory_error(mce)) + /* We only care about uncorrectable memory errors */ + if (!mce_is_memory_error(mce) || mce_is_correctable(mce)) return NOTIFY_DONE; /*

6 years, 1 month

1
0
0 0

FAILED: patch "[PATCH] acpi/nfit, x86/mce: Handle only uncorrectable machine checks" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 5d96c9342c23ee1d084802dcf064caa67ecaa45b Mon Sep 17 00:00:00 2001 From: Vishal Verma <vishal.l.verma(a)intel.com> Date: Thu, 25 Oct 2018 18:37:28 -0600 Subject: [PATCH] acpi/nfit, x86/mce: Handle only uncorrectable machine checks The MCE handler for nfit devices is called for memory errors on a Non-Volatile DIMM and adds the error location to a 'badblocks' list. This list is used by the various NVDIMM drivers to avoid consuming known poison locations during IO. The MCE handler gets called for both corrected and uncorrectable errors. Until now, both kinds of errors have been added to the badblocks list. However, corrected memory errors indicate that the problem has already been fixed by hardware, and the resulting interrupt is merely a notification to Linux. As far as future accesses to that location are concerned, it is perfectly fine to use, and thus doesn't need to be included in the above badblocks list. Add a check in the nfit MCE handler to filter out corrected mce events, and only process uncorrectable errors. Fixes: 6839a6d96f4e ("nfit: do an ARS scrub on hitting a latent media error") Reported-by: Omar Avelar <omar.avelar(a)intel.com> Signed-off-by: Vishal Verma <vishal.l.verma(a)intel.com> Signed-off-by: Borislav Petkov <bp(a)suse.de> CC: Arnd Bergmann <arnd(a)arndb.de> CC: Dan Williams <dan.j.williams(a)intel.com> CC: Dave Jiang <dave.jiang(a)intel.com> CC: elliott(a)hpe.com CC: "H. Peter Anvin" <hpa(a)zytor.com> CC: Ingo Molnar <mingo(a)redhat.com> CC: Len Brown <lenb(a)kernel.org> CC: linux-acpi(a)vger.kernel.org CC: linux-edac <linux-edac(a)vger.kernel.org> CC: linux-nvdimm(a)lists.01.org CC: Qiuxu Zhuo <qiuxu.zhuo(a)intel.com> CC: "Rafael J. Wysocki" <rjw(a)rjwysocki.net> CC: Ross Zwisler <zwisler(a)kernel.org> CC: stable <stable(a)vger.kernel.org> CC: Thomas Gleixner <tglx(a)linutronix.de> CC: Tony Luck <tony.luck(a)intel.com> CC: x86-ml <x86(a)kernel.org> CC: Yazen Ghannam <yazen.ghannam(a)amd.com> Link: http://lkml.kernel.org/r/20181026003729.8420-1-vishal.l.verma@intel.com diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 4da9b1c58d28..dbd9fe2f6163 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -221,6 +221,7 @@ static inline void mce_hygon_feature_init(struct cpuinfo_x86 *c) { return mce_am int mce_available(struct cpuinfo_x86 *c); bool mce_is_memory_error(struct mce *m); +bool mce_is_correctable(struct mce *m); DECLARE_PER_CPU(unsigned, mce_exception_count); DECLARE_PER_CPU(unsigned, mce_poll_count); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 8c66d2fc8f81..77527b8ea982 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -534,7 +534,7 @@ bool mce_is_memory_error(struct mce *m) } EXPORT_SYMBOL_GPL(mce_is_memory_error); -static bool mce_is_correctable(struct mce *m) +bool mce_is_correctable(struct mce *m) { if (m->cpuvendor == X86_VENDOR_AMD && m->status & MCI_STATUS_DEFERRED) return false; @@ -547,6 +547,7 @@ static bool mce_is_correctable(struct mce *m) return true; } +EXPORT_SYMBOL_GPL(mce_is_correctable); static bool cec_add_mce(struct mce *m) { diff --git a/drivers/acpi/nfit/mce.c b/drivers/acpi/nfit/mce.c index e9626bf6ca29..7a51707f87e9 100644 --- a/drivers/acpi/nfit/mce.c +++ b/drivers/acpi/nfit/mce.c @@ -25,8 +25,8 @@ static int nfit_handle_mce(struct notifier_block *nb, unsigned long val, struct acpi_nfit_desc *acpi_desc; struct nfit_spa *nfit_spa; - /* We only care about memory errors */ - if (!mce_is_memory_error(mce)) + /* We only care about uncorrectable memory errors */ + if (!mce_is_memory_error(mce) || mce_is_correctable(mce)) return NOTIFY_DONE; /*

6 years, 1 month

1
0
0 0

request for 4.14-stable: fec3624f0bcd ("thermal: enable broadcom menu for arm64 bcm2835")

by Sudip Mukherjee

Hi Greg, This was not marked for stable but seems it should be in stable. Please apply to your queue of 4.14-stable. -- Regards Sudip

6 years, 1 month

2
1
0 0

[PATCH stable/4.14] media: ov7670: make "xclk" clock optional

by Lubomir Rintel

[ Upstream commit 786fa584eda86d6598db3b87c61dc81f68808d11 ] When the "xclk" clock was added, it was made mandatory. This broke the driver on an OLPC plaform which doesn't know such clock. Make it optional. Tested on a OLPC XO-1 laptop. Fixes: 0a024d634cee ("[media] ov7670: get xclk") Cc: stable(a)vger.kernel.org # 4.11+ Signed-off-by: Lubomir Rintel <lkundrak(a)v3.sk> Signed-off-by: Sakari Ailus <sakari.ailus(a)linux.intel.com> Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung(a)kernel.org> --- drivers/media/i2c/ov7670.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) diff --git a/drivers/media/i2c/ov7670.c b/drivers/media/i2c/ov7670.c index e88549f0e704..39ff73a6a807 100644 --- a/drivers/media/i2c/ov7670.c +++ b/drivers/media/i2c/ov7670.c @@ -1612,23 +1612,29 @@ static int ov7670_probe(struct i2c_client *client, info->pclk_hb_disable = true; } - info->clk = devm_clk_get(&client->dev, "xclk"); - if (IS_ERR(info->clk)) - return PTR_ERR(info->clk); - ret = clk_prepare_enable(info->clk); - if (ret) - return ret; + info->clk = devm_clk_get(&client->dev, "xclk"); /* optional */ + if (IS_ERR(info->clk)) { + ret = PTR_ERR(info->clk); + if (ret == -ENOENT) + info->clk = NULL; + else + return ret; + } + if (info->clk) { + ret = clk_prepare_enable(info->clk); + if (ret) + return ret; + info->clock_speed = clk_get_rate(info->clk) / 1000000; + if (info->clock_speed < 10 || info->clock_speed > 48) { + ret = -EINVAL; + goto clk_disable; + } + } ret = ov7670_init_gpio(client, info); if (ret) goto clk_disable; - info->clock_speed = clk_get_rate(info->clk) / 1000000; - if (info->clock_speed < 10 || info->clock_speed > 48) { - ret = -EINVAL; - goto clk_disable; - } - /* Make sure it's an ov7670 */ ret = ov7670_detect(sd); if (ret) { -- 2.19.1

6 years, 1 month

2
1
0 0

request for 4.14-stable: 00c5a926af12 ("clk: mvebu: use correct bit for 98DX3236 NAND")

by Sudip Mukherjee

Hi Greg, This was not marked for stable but seems it should be in stable. Please apply to your queue of 4.14-stable. -- Regards Sudip

6 years, 1 month

2
1
0 0

FAILED: patch "[PATCH] clk: rockchip: fix wrong mmc sample phase shift for rk3328" failed to apply to 4.14-stable tree

by gregkh＠linuxfoundation.org

The patch below does not apply to the 4.14-stable tree. If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <stable(a)vger.kernel.org>. thanks, greg k-h ------------------ original commit in Linus's tree ------------------ >From 82f4b67f018c88a7cc9337f0067ed3d6ec352648 Mon Sep 17 00:00:00 2001 From: Ziyuan Xu <xzy.xu(a)rock-chips.com> Date: Thu, 11 Oct 2018 15:26:43 +0800 Subject: [PATCH] clk: rockchip: fix wrong mmc sample phase shift for rk3328 mmc sample shift is 0 for RK3328 referring to the TRM. So fix them. Fixes: fe3511ad8a1c ("clk: rockchip: add clock controller for rk3328") Cc: stable(a)vger.kernel.org Signed-off-by: Ziyuan Xu <xzy.xu(a)rock-chips.com> Signed-off-by: Shawn Lin <shawn.lin(a)rock-chips.com> Signed-off-by: Heiko Stuebner <heiko(a)sntech.de> diff --git a/drivers/clk/rockchip/clk-rk3328.c b/drivers/clk/rockchip/clk-rk3328.c index 252366a5231f..2c5426607790 100644 --- a/drivers/clk/rockchip/clk-rk3328.c +++ b/drivers/clk/rockchip/clk-rk3328.c @@ -813,22 +813,22 @@ static struct rockchip_clk_branch rk3328_clk_branches[] __initdata = { MMC(SCLK_SDMMC_DRV, "sdmmc_drv", "clk_sdmmc", RK3328_SDMMC_CON0, 1), MMC(SCLK_SDMMC_SAMPLE, "sdmmc_sample", "clk_sdmmc", - RK3328_SDMMC_CON1, 1), + RK3328_SDMMC_CON1, 0), MMC(SCLK_SDIO_DRV, "sdio_drv", "clk_sdio", RK3328_SDIO_CON0, 1), MMC(SCLK_SDIO_SAMPLE, "sdio_sample", "clk_sdio", - RK3328_SDIO_CON1, 1), + RK3328_SDIO_CON1, 0), MMC(SCLK_EMMC_DRV, "emmc_drv", "clk_emmc", RK3328_EMMC_CON0, 1), MMC(SCLK_EMMC_SAMPLE, "emmc_sample", "clk_emmc", - RK3328_EMMC_CON1, 1), + RK3328_EMMC_CON1, 0), MMC(SCLK_SDMMC_EXT_DRV, "sdmmc_ext_drv", "clk_sdmmc_ext", RK3328_SDMMC_EXT_CON0, 1), MMC(SCLK_SDMMC_EXT_SAMPLE, "sdmmc_ext_sample", "clk_sdmmc_ext", - RK3328_SDMMC_EXT_CON1, 1), + RK3328_SDMMC_EXT_CON1, 0), }; static const char *const rk3328_critical_clocks[] __initconst = {

6 years, 1 month

1
0
0 0

[PATCH V2 for-4.21 1/2] blk-mq: not embed .mq_kobj and ctx->kobj into queue instance

by Ming Lei

Even though .mq_kobj, ctx->kobj and q->kobj share same lifetime from block layer's view, actually they don't because userspace may grab one kobject anytime via sysfs, so each kobject's lifetime has to be independent, then the objects(mq_kobj, ctx) which hosts its own kobject have to be allocated dynamically. This patch fixes kernel panic issue during booting when DEBUG_KOBJECT_RELEASE is enabled. Reported-by: Guenter Roeck <linux(a)roeck-us.net> Cc: "jianchao.wang" <jianchao.w.wang(a)oracle.com> Cc: Guenter Roeck <linux(a)roeck-us.net> Cc: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Cc: stable(a)vger.kernel.org Signed-off-by: Ming Lei <ming.lei(a)redhat.com> --- block/blk-mq-sysfs.c | 41 +++++++++++++++++++++++++------------ block/blk-mq.c | 55 +++++++++++++++++++++++++++++++++++++++++--------- block/blk-mq.h | 4 ++-- include/linux/blkdev.h | 4 ++-- 4 files changed, 78 insertions(+), 26 deletions(-) diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c index 3d25b9c419e9..cc2fef909afc 100644 --- a/block/blk-mq-sysfs.c +++ b/block/blk-mq-sysfs.c @@ -15,6 +15,14 @@ static void blk_mq_sysfs_release(struct kobject *kobj) { + kfree(kobj); +} + +static void blk_mq_ctx_sysfs_release(struct kobject *kobj) +{ + struct blk_mq_ctx *ctx = container_of(kobj, struct blk_mq_ctx, kobj); + + kfree(ctx); } static void blk_mq_hw_sysfs_release(struct kobject *kobj) @@ -213,7 +221,7 @@ static struct kobj_type blk_mq_ktype = { static struct kobj_type blk_mq_ctx_ktype = { .sysfs_ops = &blk_mq_sysfs_ops, .default_attrs = default_ctx_attrs, - .release = blk_mq_sysfs_release, + .release = blk_mq_ctx_sysfs_release, }; static struct kobj_type blk_mq_hw_ktype = { @@ -245,7 +253,7 @@ static int blk_mq_register_hctx(struct blk_mq_hw_ctx *hctx) if (!hctx->nr_ctx) return 0; - ret = kobject_add(&hctx->kobj, &q->mq_kobj, "%u", hctx->queue_num); + ret = kobject_add(&hctx->kobj, q->mq_kobj, "%u", hctx->queue_num); if (ret) return ret; @@ -268,8 +276,8 @@ void blk_mq_unregister_dev(struct device *dev, struct request_queue *q) queue_for_each_hw_ctx(q, hctx, i) blk_mq_unregister_hctx(hctx); - kobject_uevent(&q->mq_kobj, KOBJ_REMOVE); - kobject_del(&q->mq_kobj); + kobject_uevent(q->mq_kobj, KOBJ_REMOVE); + kobject_del(q->mq_kobj); kobject_put(&dev->kobj); q->mq_sysfs_init_done = false; @@ -286,23 +294,30 @@ void blk_mq_sysfs_deinit(struct request_queue *q) int cpu; for_each_possible_cpu(cpu) { - ctx = per_cpu_ptr(q->queue_ctx, cpu); + ctx = *per_cpu_ptr(q->queue_ctx, cpu); kobject_put(&ctx->kobj); } - kobject_put(&q->mq_kobj); + kobject_put(q->mq_kobj); } -void blk_mq_sysfs_init(struct request_queue *q) +int blk_mq_sysfs_init(struct request_queue *q) { struct blk_mq_ctx *ctx; int cpu; + struct kobject *mq_kobj; + + mq_kobj = kzalloc(sizeof(*mq_kobj), GFP_KERNEL); + if (!mq_kobj) + return -ENOMEM; - kobject_init(&q->mq_kobj, &blk_mq_ktype); + kobject_init(mq_kobj, &blk_mq_ktype); for_each_possible_cpu(cpu) { - ctx = per_cpu_ptr(q->queue_ctx, cpu); + ctx = *per_cpu_ptr(q->queue_ctx, cpu); kobject_init(&ctx->kobj, &blk_mq_ctx_ktype); } + q->mq_kobj = mq_kobj; + return 0; } int __blk_mq_register_dev(struct device *dev, struct request_queue *q) @@ -313,11 +328,11 @@ int __blk_mq_register_dev(struct device *dev, struct request_queue *q) WARN_ON_ONCE(!q->kobj.parent); lockdep_assert_held(&q->sysfs_lock); - ret = kobject_add(&q->mq_kobj, kobject_get(&dev->kobj), "%s", "mq"); + ret = kobject_add(q->mq_kobj, kobject_get(&dev->kobj), "%s", "mq"); if (ret < 0) goto out; - kobject_uevent(&q->mq_kobj, KOBJ_ADD); + kobject_uevent(q->mq_kobj, KOBJ_ADD); queue_for_each_hw_ctx(q, hctx, i) { ret = blk_mq_register_hctx(hctx); @@ -334,8 +349,8 @@ int __blk_mq_register_dev(struct device *dev, struct request_queue *q) while (--i >= 0) blk_mq_unregister_hctx(q->queue_hw_ctx[i]); - kobject_uevent(&q->mq_kobj, KOBJ_REMOVE); - kobject_del(&q->mq_kobj); + kobject_uevent(q->mq_kobj, KOBJ_REMOVE); + kobject_del(q->mq_kobj); kobject_put(&dev->kobj); return ret; } diff --git a/block/blk-mq.c b/block/blk-mq.c index 3b823891b3ef..376c04778d33 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -2299,7 +2299,7 @@ static void blk_mq_init_cpu_queues(struct request_queue *q, unsigned int i, j; for_each_possible_cpu(i) { - struct blk_mq_ctx *__ctx = per_cpu_ptr(q->queue_ctx, i); + struct blk_mq_ctx *__ctx = *per_cpu_ptr(q->queue_ctx, i); struct blk_mq_hw_ctx *hctx; __ctx->cpu = i; @@ -2385,7 +2385,7 @@ static void blk_mq_map_swqueue(struct request_queue *q) set->map[0].mq_map[i] = 0; } - ctx = per_cpu_ptr(q->queue_ctx, i); + ctx = *per_cpu_ptr(q->queue_ctx, i); for (j = 0; j < set->nr_maps; j++) { hctx = blk_mq_map_queue_type(q, j, i); @@ -2515,6 +2515,38 @@ static void blk_mq_add_queue_tag_set(struct blk_mq_tag_set *set, mutex_unlock(&set->tag_list_lock); } +static void blk_mq_dealloc_queue_ctx(struct request_queue *q, bool free_ctxs) +{ + if (free_ctxs) { + int cpu; + for_each_possible_cpu(cpu) + kfree(*per_cpu_ptr(q->queue_ctx, cpu)); + } + free_percpu(q->queue_ctx); +} + +static int blk_mq_alloc_queue_ctx(struct request_queue *q) +{ + struct blk_mq_ctx *ctx; + int cpu; + + q->queue_ctx = alloc_percpu(struct blk_mq_ctx *); + if (!q->queue_ctx) + return -ENOMEM; + + for_each_possible_cpu(cpu) { + ctx = kzalloc_node(sizeof(*ctx), GFP_KERNEL, cpu_to_node(cpu)); + if (!ctx) + goto fail; + *per_cpu_ptr(q->queue_ctx, cpu) = ctx; + } + + return 0; + fail: + blk_mq_dealloc_queue_ctx(q, true); + return -ENOMEM; +} + /* * It is the actual release handler for mq, but we do it from * request queue's release handler for avoiding use-after-free @@ -2541,7 +2573,7 @@ void blk_mq_release(struct request_queue *q) */ blk_mq_sysfs_deinit(q); - free_percpu(q->queue_ctx); + blk_mq_dealloc_queue_ctx(q, false); } struct request_queue *blk_mq_init_queue(struct blk_mq_tag_set *set) @@ -2722,6 +2754,8 @@ static unsigned int nr_hw_queues(struct blk_mq_tag_set *set) struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, struct request_queue *q) { + bool sysfs_init_done = false; + /* mark the queue as mq asap */ q->mq_ops = set->ops; @@ -2731,18 +2765,19 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, if (!q->poll_cb) goto err_exit; - q->queue_ctx = alloc_percpu(struct blk_mq_ctx); - if (!q->queue_ctx) + if (blk_mq_alloc_queue_ctx(q)) goto err_exit; /* init q->mq_kobj and sw queues' kobjects */ - blk_mq_sysfs_init(q); + if (blk_mq_sysfs_init(q)) + goto err_queue_ctx; + sysfs_init_done = true; q->nr_queues = nr_hw_queues(set); q->queue_hw_ctx = kcalloc_node(q->nr_queues, sizeof(*(q->queue_hw_ctx)), GFP_KERNEL, set->numa_node); if (!q->queue_hw_ctx) - goto err_percpu; + goto err_sys_init; blk_mq_realloc_hw_ctxs(set, q); if (!q->nr_hw_queues) @@ -2794,8 +2829,10 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, err_hctxs: kfree(q->queue_hw_ctx); -err_percpu: - free_percpu(q->queue_ctx); +err_sys_init: + blk_mq_sysfs_deinit(q); +err_queue_ctx: + blk_mq_dealloc_queue_ctx(q, !sysfs_init_done); err_exit: q->mq_ops = NULL; return ERR_PTR(-ENOMEM); diff --git a/block/blk-mq.h b/block/blk-mq.h index facb6e9ddce4..84898793c230 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -108,7 +108,7 @@ static inline struct blk_mq_hw_ctx *blk_mq_map_queue(struct request_queue *q, /* * sysfs helpers */ -extern void blk_mq_sysfs_init(struct request_queue *q); +extern int blk_mq_sysfs_init(struct request_queue *q); extern void blk_mq_sysfs_deinit(struct request_queue *q); extern int __blk_mq_register_dev(struct device *dev, struct request_queue *q); extern int blk_mq_sysfs_register(struct request_queue *q); @@ -129,7 +129,7 @@ static inline enum mq_rq_state blk_mq_rq_state(struct request *rq) static inline struct blk_mq_ctx *__blk_mq_get_ctx(struct request_queue *q, unsigned int cpu) { - return per_cpu_ptr(q->queue_ctx, cpu); + return *per_cpu_ptr(q->queue_ctx, cpu); } /* diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 1d185f1fc333..9e3892bd67fd 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -407,7 +407,7 @@ struct request_queue { const struct blk_mq_ops *mq_ops; /* sw queues */ - struct blk_mq_ctx __percpu *queue_ctx; + struct blk_mq_ctx __percpu **queue_ctx; unsigned int nr_queues; unsigned int queue_depth; @@ -456,7 +456,7 @@ struct request_queue { /* * mq queue kobject */ - struct kobject mq_kobj; + struct kobject *mq_kobj; #ifdef CONFIG_BLK_DEV_INTEGRITY struct blk_integrity integrity; -- 2.9.5

6 years, 1 month

2
4
0 0

[PATCH] Revert "usb: dwc3: gadget: skip Set/Clear Halt when invalid"

by Felipe Balbi

This reverts commit ffb80fc672c3a7b6afd0cefcb1524fb99917b2f3. Turns out that commit is wrong. Host controllers are allowed to use Clear Feature HALT as means to sync data toggle between host and periperal. Cc: <stable(a)vger.kernel.org> Signed-off-by: Felipe Balbi <felipe.balbi(a)linux.intel.com> --- drivers/usb/dwc3/gadget.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index 9faad896b3a1..9f92ee03dde7 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -1470,9 +1470,6 @@ int __dwc3_gadget_ep_set_halt(struct dwc3_ep *dep, int value, int protocol) unsigned transfer_in_flight; unsigned started; - if (dep->flags & DWC3_EP_STALL) - return 0; - if (dep->number > 1) trb = dwc3_ep_prev_trb(dep, dep->trb_enqueue); else @@ -1494,8 +1491,6 @@ int __dwc3_gadget_ep_set_halt(struct dwc3_ep *dep, int value, int protocol) else dep->flags |= DWC3_EP_STALL; } else { - if (!(dep->flags & DWC3_EP_STALL)) - return 0; ret = dwc3_send_clear_stall_ep_cmd(dep); if (ret) -- 2.19.1

6 years, 1 month

1
0
0 0

Re: make a confirm for [usb: dwc3: gadget: skip Set/Clear Halt when invalid]

by Felipe Balbi

Hi, that patch is not 100% correct. You can revert it in your tree. I added that because of a problem I found when running adb against macOS. It's actually okay to send Clear Halt at any time, but for some reason dwc3 was hanging when running adb against macOS. If you can revert the patch and make sure it works against all 3 major OSes (linux, windows and mac) I'd be really glad. liangshengjun <liangshengjun(a)hisilicon.com> writes: > Hi felipe, > > I have met a case about set/clear Halt patch > Version: linux v4.16, > Case: usb uvc run with bulk-mode connect to Windows 7 PC. When PC stop camera application , it would send clearHalt request to uvc device to streaming-off video transfer. > But with v4.16 dwc3 drivers, it would skip handling this clear Halt request , because dep->flags is not DWC3_EP_STALL status, then it causes PC restart camera application , uvc transfer fail. > And I have confirmed v3.18 dwc3 drivers is OK. > > So how to balance for handling clear Halt without first setHalt ?? > > PS: > commit ffb80fc672c3a7b6afd0cefcb1524fb99917b2f3 > Author: Felipe Balbi <felipe.balbi(a)linux.intel.com> > Date: Thu Jan 19 13:38:42 2017 +0200 > > usb: dwc3: gadget: skip Set/Clear Halt when invalid > > At least macOS seems to be sending > ClearFeature(ENDPOINT_HALT) to endpoints which > aren't Halted. This makes DWC3's CLEARSTALL command > time out which causes several issues for the driver. > > Instead, let's just return 0 and bail out early. > > Cc: <stable(a)vger.kernel.org> > Signed-off-by: Felipe Balbi <felipe.balbi(a)linux.intel.com> > > diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c > index 6faf484..0a664d8 100644 > --- a/drivers/usb/dwc3/gadget.c > +++ b/drivers/usb/dwc3/gadget.c > @@ -1379,6 +1379,9 @@ int __dwc3_gadget_ep_set_halt(struct dwc3_ep *dep, int value, int protocol) > unsigned transfer_in_flight; > unsigned started; > > + if (dep->flags & DWC3_EP_STALL) > + return 0; > + > if (dep->number > 1) > trb = dwc3_ep_prev_trb(dep, dep->trb_enqueue); > else > @@ -1400,6 +1403,8 @@ int __dwc3_gadget_ep_set_halt(struct dwc3_ep *dep, int value, int protocol) > else > dep->flags |= DWC3_EP_STALL; > } else { > + if (!(dep->flags & DWC3_EP_STALL)) > + return 0; > > ret = dwc3_send_clear_stall_ep_cmd(dep); > if (ret) > > > Liang Shengjun > [cid:image001.png@01D40971.9265B340] > HISILICON TECHNOLOGIES CO., LTD. > New R&D Center, Wuhe Road, Bantian, > Longgang District, Shenzhen 518129 P.R. China > -- balbi

6 years, 1 month

4
6
0 0

Request for 3.18-stable: a63b03e2d247 ("mutex: Always clear owner field upon mutex_unlock()")

by Tom Zanussi

Hi Greg, I noticed that 3.18.125 added commit bc07ee33284a ('Revert "drm/i915: Fix mutex->owner inspection race under DEBUG_MUTEXES"'), which states that the reason it can be applied is: The core fix was applied in commit a63b03e2d2477586440741677ecac45bcf28d7b1 Author: Chris Wilson <chris(a)chris-wilson.co.uk> Date: Tue Jan 6 10:29:35 2015 +0000 mutex: Always clear owner field upon mutex_unlock() (note the absence of stable@ tag) so we can now revert our band-aid commit 226e5ae9e5f910 for -next. but that the commit referenced wasn't also pulled in. Please consider pulling that one too if you're going to do another 3.18 stable release. Thanks, Tom

6 years, 1 month

2
2
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror November 2018