From: Sasha Levin Date: Tue, 10 Jan 2023 01:55:37 +0000 (-0500) Subject: Fixes for 5.4 X-Git-Tag: v5.15.87~42 X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=b1dc6e206e4974fca572325cd157fea95357926b;p=thirdparty%2Fkernel%2Fstable-queue.git Fixes for 5.4 Signed-off-by: Sasha Levin --- diff --git a/queue-5.4/asoc-intel-bytcr_rt5640-add-quirk-for-the-advantech-.patch b/queue-5.4/asoc-intel-bytcr_rt5640-add-quirk-for-the-advantech-.patch new file mode 100644 index 00000000000..4a6ab6a9607 --- /dev/null +++ b/queue-5.4/asoc-intel-bytcr_rt5640-add-quirk-for-the-advantech-.patch @@ -0,0 +1,59 @@ +From 519054b3d9d59cf90ea518d6ad6a11f05fb35b3e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 13 Dec 2022 13:32:46 +0100 +Subject: ASoC: Intel: bytcr_rt5640: Add quirk for the Advantech MICA-071 + tablet + +From: Hans de Goede + +[ Upstream commit a1dec9d70b6ad97087b60b81d2492134a84208c6 ] + +The Advantech MICA-071 tablet deviates from the defaults for +a non CR Bay Trail based tablet in several ways: + +1. It uses an analog MIC on IN3 rather then using DMIC1 +2. It only has 1 speaker +3. It needs the OVCD current threshold to be set to 1500uA instead of + the default 2000uA to reliable differentiate between headphones vs + headsets + +Add a quirk with these settings for this tablet. + +Signed-off-by: Hans de Goede +Acked-by: Pierre-Louis Bossart +Link: https://lore.kernel.org/r/20221213123246.11226-1-hdegoede@redhat.com +Signed-off-by: Mark Brown +Signed-off-by: Sasha Levin +--- + sound/soc/intel/boards/bytcr_rt5640.c | 15 +++++++++++++++ + 1 file changed, 15 insertions(+) + +diff --git a/sound/soc/intel/boards/bytcr_rt5640.c b/sound/soc/intel/boards/bytcr_rt5640.c +index 7830d014d924..6a8edb0a559d 100644 +--- a/sound/soc/intel/boards/bytcr_rt5640.c ++++ b/sound/soc/intel/boards/bytcr_rt5640.c +@@ -428,6 +428,21 @@ static const struct dmi_system_id byt_rt5640_quirk_table[] = { + BYT_RT5640_SSP0_AIF1 | + BYT_RT5640_MCLK_EN), + }, ++ { ++ /* Advantech MICA-071 */ ++ .matches = { ++ DMI_EXACT_MATCH(DMI_SYS_VENDOR, "Advantech"), ++ DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "MICA-071"), ++ }, ++ /* OVCD Th = 1500uA to reliable detect head-phones vs -set */ ++ .driver_data = (void *)(BYT_RT5640_IN3_MAP | ++ BYT_RT5640_JD_SRC_JD2_IN4N | ++ BYT_RT5640_OVCD_TH_1500UA | ++ BYT_RT5640_OVCD_SF_0P75 | ++ BYT_RT5640_MONO_SPEAKER | ++ BYT_RT5640_DIFF_MIC | ++ BYT_RT5640_MCLK_EN), ++ }, + { + .matches = { + DMI_EXACT_MATCH(DMI_SYS_VENDOR, "ARCHOS"), +-- +2.35.1 + diff --git a/queue-5.4/bpf-pull-before-calling-skb_postpull_rcsum.patch b/queue-5.4/bpf-pull-before-calling-skb_postpull_rcsum.patch new file mode 100644 index 00000000000..5a894b57ac5 --- /dev/null +++ b/queue-5.4/bpf-pull-before-calling-skb_postpull_rcsum.patch @@ -0,0 +1,61 @@ +From e8522a77d18724fa5fbc00e97a697aa6264d3e14 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 19 Dec 2022 16:47:00 -0800 +Subject: bpf: pull before calling skb_postpull_rcsum() + +From: Jakub Kicinski + +[ Upstream commit 54c3f1a81421f85e60ae2eaae7be3727a09916ee ] + +Anand hit a BUG() when pulling off headers on egress to a SW tunnel. +We get to skb_checksum_help() with an invalid checksum offset +(commit d7ea0d9df2a6 ("net: remove two BUG() from skb_checksum_help()") +converted those BUGs to WARN_ONs()). +He points out oddness in how skb_postpull_rcsum() gets used. +Indeed looks like we should pull before "postpull", otherwise +the CHECKSUM_PARTIAL fixup from skb_postpull_rcsum() will not +be able to do its job: + + if (skb->ip_summed == CHECKSUM_PARTIAL && + skb_checksum_start_offset(skb) < 0) + skb->ip_summed = CHECKSUM_NONE; + +Reported-by: Anand Parthasarathy +Fixes: 6578171a7ff0 ("bpf: add bpf_skb_change_proto helper") +Signed-off-by: Jakub Kicinski +Acked-by: Stanislav Fomichev +Link: https://lore.kernel.org/r/20221220004701.402165-1-kuba@kernel.org +Signed-off-by: Martin KaFai Lau +Signed-off-by: Sasha Levin +--- + net/core/filter.c | 7 +++++-- + 1 file changed, 5 insertions(+), 2 deletions(-) + +diff --git a/net/core/filter.c b/net/core/filter.c +index ec4f7e68b21a..71fcb4e7edae 100644 +--- a/net/core/filter.c ++++ b/net/core/filter.c +@@ -2795,15 +2795,18 @@ static int bpf_skb_generic_push(struct sk_buff *skb, u32 off, u32 len) + + static int bpf_skb_generic_pop(struct sk_buff *skb, u32 off, u32 len) + { ++ void *old_data; ++ + /* skb_ensure_writable() is not needed here, as we're + * already working on an uncloned skb. + */ + if (unlikely(!pskb_may_pull(skb, off + len))) + return -ENOMEM; + +- skb_postpull_rcsum(skb, skb->data + off, len); +- memmove(skb->data + len, skb->data, off); ++ old_data = skb->data; + __skb_pull(skb, len); ++ skb_postpull_rcsum(skb, old_data + off, len); ++ memmove(skb->data, old_data, off); + + return 0; + } +-- +2.35.1 + diff --git a/queue-5.4/caif-fix-memory-leak-in-cfctrl_linkup_request.patch b/queue-5.4/caif-fix-memory-leak-in-cfctrl_linkup_request.patch new file mode 100644 index 00000000000..1198b855700 --- /dev/null +++ b/queue-5.4/caif-fix-memory-leak-in-cfctrl_linkup_request.patch @@ -0,0 +1,47 @@ +From bc57e7edd2dee1b788d8192eb049fca24c4bab97 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 4 Jan 2023 14:51:46 +0800 +Subject: caif: fix memory leak in cfctrl_linkup_request() + +From: Zhengchao Shao + +[ Upstream commit fe69230f05897b3de758427b574fc98025dfc907 ] + +When linktype is unknown or kzalloc failed in cfctrl_linkup_request(), +pkt is not released. Add release process to error path. + +Fixes: b482cd2053e3 ("net-caif: add CAIF core protocol stack") +Fixes: 8d545c8f958f ("caif: Disconnect without waiting for response") +Signed-off-by: Zhengchao Shao +Reviewed-by: Jiri Pirko +Link: https://lore.kernel.org/r/20230104065146.1153009-1-shaozhengchao@huawei.com +Signed-off-by: Paolo Abeni +Signed-off-by: Sasha Levin +--- + net/caif/cfctrl.c | 6 +++++- + 1 file changed, 5 insertions(+), 1 deletion(-) + +diff --git a/net/caif/cfctrl.c b/net/caif/cfctrl.c +index 2809cbd6b7f7..d8cb4b2a076b 100644 +--- a/net/caif/cfctrl.c ++++ b/net/caif/cfctrl.c +@@ -269,11 +269,15 @@ int cfctrl_linkup_request(struct cflayer *layer, + default: + pr_warn("Request setup of bad link type = %d\n", + param->linktype); ++ cfpkt_destroy(pkt); + return -EINVAL; + } + req = kzalloc(sizeof(*req), GFP_KERNEL); +- if (!req) ++ if (!req) { ++ cfpkt_destroy(pkt); + return -ENOMEM; ++ } ++ + req->client_layer = user_layer; + req->cmd = CFCTRL_CMD_LINK_SETUP; + req->param = *param; +-- +2.35.1 + diff --git a/queue-5.4/dm-thin-resume-even-if-in-fail-mode.patch b/queue-5.4/dm-thin-resume-even-if-in-fail-mode.patch new file mode 100644 index 00000000000..0ee8362fb36 --- /dev/null +++ b/queue-5.4/dm-thin-resume-even-if-in-fail-mode.patch @@ -0,0 +1,80 @@ +From ac99413d3a22b5c9db7fc55f0ecca8d9798a622f Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 30 Nov 2022 10:09:45 +0800 +Subject: dm thin: resume even if in FAIL mode + +From: Luo Meng + +[ Upstream commit 19eb1650afeb1aa86151f61900e9e5f1de5d8d02 ] + +If a thinpool set fail_io while suspending, resume will fail with: + device-mapper: resume ioctl on vg-thinpool failed: Invalid argument + +The thin-pool also can't be removed if an in-flight bio is in the +deferred list. + +This can be easily reproduced using: + + echo "offline" > /sys/block/sda/device/state + dd if=/dev/zero of=/dev/mapper/thin bs=4K count=1 + dmsetup suspend /dev/mapper/pool + mkfs.ext4 /dev/mapper/thin + dmsetup resume /dev/mapper/pool + +The root cause is maybe_resize_data_dev() will check fail_io and return +error before called dm_resume. + +Fix this by adding FAIL mode check at the end of pool_preresume(). + +Cc: stable@vger.kernel.org +Fixes: da105ed5fd7e ("dm thin metadata: introduce dm_pool_abort_metadata") +Signed-off-by: Luo Meng +Signed-off-by: Mike Snitzer +Signed-off-by: Sasha Levin +--- + drivers/md/dm-thin.c | 16 ++++++++++++---- + 1 file changed, 12 insertions(+), 4 deletions(-) + +diff --git a/drivers/md/dm-thin.c b/drivers/md/dm-thin.c +index 1af5873923e8..4f161725dda0 100644 +--- a/drivers/md/dm-thin.c ++++ b/drivers/md/dm-thin.c +@@ -3593,23 +3593,31 @@ static int pool_preresume(struct dm_target *ti) + */ + r = bind_control_target(pool, ti); + if (r) +- return r; ++ goto out; + + dm_pool_register_pre_commit_callback(pool->pmd, + metadata_pre_commit_callback, pt); + + r = maybe_resize_data_dev(ti, &need_commit1); + if (r) +- return r; ++ goto out; + + r = maybe_resize_metadata_dev(ti, &need_commit2); + if (r) +- return r; ++ goto out; + + if (need_commit1 || need_commit2) + (void) commit(pool); ++out: ++ /* ++ * When a thin-pool is PM_FAIL, it cannot be rebuilt if ++ * bio is in deferred list. Therefore need to return 0 ++ * to allow pool_resume() to flush IO. ++ */ ++ if (r && get_pool_mode(pool) == PM_FAIL) ++ r = 0; + +- return 0; ++ return r; + } + + static void pool_suspend_active_thins(struct pool *pool) +-- +2.35.1 + diff --git a/queue-5.4/driver-core-fix-driver_deferred_probe_check_state-lo.patch b/queue-5.4/driver-core-fix-driver_deferred_probe_check_state-lo.patch new file mode 100644 index 00000000000..0bbbac271f7 --- /dev/null +++ b/queue-5.4/driver-core-fix-driver_deferred_probe_check_state-lo.patch @@ -0,0 +1,124 @@ +From e507b1ffd90ffc541fe254be4b63d4f3ff420a7c Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 25 Feb 2020 05:08:23 +0000 +Subject: driver core: Fix driver_deferred_probe_check_state() logic + +From: John Stultz + +[ Upstream commit c8c43cee29f6ca2575c953ae600263690db28f41 ] + +driver_deferred_probe_check_state() has some uninituitive behavior. + +* From boot to late_initcall, it returns -EPROBE_DEFER + +* From late_initcall to the deferred_probe_timeout (if set) + it returns -ENODEV + +* If the deferred_probe_timeout it set, after it fires, it + returns -ETIMEDOUT + +This is a bit confusing, as its useful to have the function +return -EPROBE_DEFER while the timeout is still running. This +behavior has resulted in the somwhat duplicative +driver_deferred_probe_check_state_continue() function being +added. + +Thus this patch tries to improve the logic, so that it behaves +as such: + +* If late_initcall has passed, and modules are not enabled + it returns -ENODEV + +* If modules are enabled and deferred_probe_timeout is set, + it returns -EPROBE_DEFER until the timeout, afterwhich it + returns -ETIMEDOUT. + +* In all other cases, it returns -EPROBE_DEFER + +This will make the deferred_probe_timeout value much more +functional, and will allow us to consolidate the +driver_deferred_probe_check_state() and +driver_deferred_probe_check_state_continue() logic in a later +patch. + +Cc: linux-pm@vger.kernel.org +Cc: Greg Kroah-Hartman +Cc: Linus Walleij +Cc: Thierry Reding +Cc: Mark Brown +Cc: Liam Girdwood +Cc: Bjorn Andersson +Cc: Saravana Kannan +Cc: Todd Kjos +Cc: Len Brown +Cc: Pavel Machek +Cc: Ulf Hansson +Cc: Kevin Hilman +Cc: "Rafael J. Wysocki" +Cc: Rob Herring +Signed-off-by: John Stultz +Link: https://lore.kernel.org/r/20200225050828.56458-2-john.stultz@linaro.org +Signed-off-by: Greg Kroah-Hartman +Signed-off-by: Sasha Levin +--- + drivers/base/dd.c | 18 ++++++++++-------- + 1 file changed, 10 insertions(+), 8 deletions(-) + +diff --git a/drivers/base/dd.c b/drivers/base/dd.c +index 10063d8a1b7d..2a28e07d2e0b 100644 +--- a/drivers/base/dd.c ++++ b/drivers/base/dd.c +@@ -237,24 +237,26 @@ __setup("deferred_probe_timeout=", deferred_probe_timeout_setup); + + static int __driver_deferred_probe_check_state(struct device *dev) + { +- if (!initcalls_done) +- return -EPROBE_DEFER; ++ if (!IS_ENABLED(CONFIG_MODULES) && initcalls_done) ++ return -ENODEV; + + if (!deferred_probe_timeout) { + dev_WARN(dev, "deferred probe timeout, ignoring dependency"); + return -ETIMEDOUT; + } + +- return 0; ++ return -EPROBE_DEFER; + } + + /** + * driver_deferred_probe_check_state() - Check deferred probe state + * @dev: device to check + * +- * Returns -ENODEV if init is done and all built-in drivers have had a chance +- * to probe (i.e. initcalls are done), -ETIMEDOUT if deferred probe debug +- * timeout has expired, or -EPROBE_DEFER if none of those conditions are met. ++ * Return: ++ * -ENODEV if initcalls have completed and modules are disabled. ++ * -ETIMEDOUT if the deferred probe timeout was set and has expired ++ * and modules are enabled. ++ * -EPROBE_DEFER in other cases. + * + * Drivers or subsystems can opt-in to calling this function instead of directly + * returning -EPROBE_DEFER. +@@ -264,7 +266,7 @@ int driver_deferred_probe_check_state(struct device *dev) + int ret; + + ret = __driver_deferred_probe_check_state(dev); +- if (ret < 0) ++ if (ret != -ENODEV) + return ret; + + dev_warn(dev, "ignoring dependency for device, assuming no driver"); +@@ -292,7 +294,7 @@ int driver_deferred_probe_check_state_continue(struct device *dev) + int ret; + + ret = __driver_deferred_probe_check_state(dev); +- if (ret < 0) ++ if (ret != -ENODEV) + return ret; + + return -EPROBE_DEFER; +-- +2.35.1 + diff --git a/queue-5.4/driver-core-set-deferred_probe_timeout-to-a-longer-d.patch b/queue-5.4/driver-core-set-deferred_probe_timeout-to-a-longer-d.patch new file mode 100644 index 00000000000..2553d44655a --- /dev/null +++ b/queue-5.4/driver-core-set-deferred_probe_timeout-to-a-longer-d.patch @@ -0,0 +1,69 @@ +From cae2bae22ae13e0c31293862dc9addea3118695e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 25 Feb 2020 05:08:24 +0000 +Subject: driver core: Set deferred_probe_timeout to a longer default if + CONFIG_MODULES is set + +From: John Stultz + +[ Upstream commit e2cec7d6853712295cef5377762165a489b2957f ] + +When using modules, its common for the modules not to be loaded +until quite late by userland. With the current code, +driver_deferred_probe_check_state() will stop returning +EPROBE_DEFER after late_initcall, which can cause module +dependency resolution to fail after that. + +So allow a longer window of 30 seconds (picked somewhat +arbitrarily, but influenced by the similar regulator core +timeout value) in the case where modules are enabled. + +Cc: linux-pm@vger.kernel.org +Cc: Greg Kroah-Hartman +Cc: Linus Walleij +Cc: Thierry Reding +Cc: Mark Brown +Cc: Liam Girdwood +Cc: Bjorn Andersson +Cc: Saravana Kannan +Cc: Todd Kjos +Cc: Len Brown +Cc: Pavel Machek +Cc: Ulf Hansson +Cc: Kevin Hilman +Cc: "Rafael J. Wysocki" +Cc: Rob Herring +Reviewed-by: Bjorn Andersson +Reviewed-by: Rafael J. Wysocki +Signed-off-by: John Stultz +Link: https://lore.kernel.org/r/20200225050828.56458-3-john.stultz@linaro.org +Signed-off-by: Greg Kroah-Hartman +Signed-off-by: Sasha Levin +--- + drivers/base/dd.c | 9 +++++++++ + 1 file changed, 9 insertions(+) + +diff --git a/drivers/base/dd.c b/drivers/base/dd.c +index 2a28e07d2e0b..490cffffcae4 100644 +--- a/drivers/base/dd.c ++++ b/drivers/base/dd.c +@@ -224,7 +224,16 @@ static int deferred_devs_show(struct seq_file *s, void *data) + } + DEFINE_SHOW_ATTRIBUTE(deferred_devs); + ++#ifdef CONFIG_MODULES ++/* ++ * In the case of modules, set the default probe timeout to ++ * 30 seconds to give userland some time to load needed modules ++ */ ++static int deferred_probe_timeout = 30; ++#else ++/* In the case of !modules, no probe timeout needed */ + static int deferred_probe_timeout = -1; ++#endif + static int __init deferred_probe_timeout_setup(char *str) + { + int timeout; +-- +2.35.1 + diff --git a/queue-5.4/drivers-net-bonding-bond_3ad-return-when-there-s-no-.patch b/queue-5.4/drivers-net-bonding-bond_3ad-return-when-there-s-no-.patch new file mode 100644 index 00000000000..5baec0503c6 --- /dev/null +++ b/queue-5.4/drivers-net-bonding-bond_3ad-return-when-there-s-no-.patch @@ -0,0 +1,39 @@ +From 957709588ecf1922dd0ea67c5aa8d5c7a71c5e6d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 2 Jan 2023 12:53:35 +0300 +Subject: drivers/net/bonding/bond_3ad: return when there's no aggregator + +From: Daniil Tatianin + +[ Upstream commit 9c807965483f42df1d053b7436eedd6cf28ece6f ] + +Otherwise we would dereference a NULL aggregator pointer when calling +__set_agg_ports_ready on the line below. + +Found by Linux Verification Center (linuxtesting.org) with the SVACE +static analysis tool. + +Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") +Signed-off-by: Daniil Tatianin +Reviewed-by: Jiri Pirko +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + drivers/net/bonding/bond_3ad.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c +index 0d6cd2a4cc41..0c4e6fcac58e 100644 +--- a/drivers/net/bonding/bond_3ad.c ++++ b/drivers/net/bonding/bond_3ad.c +@@ -1529,6 +1529,7 @@ static void ad_port_selection_logic(struct port *port, bool *update_slave_arr) + slave_err(bond->dev, port->slave->dev, + "Port %d did not find a suitable aggregator\n", + port->actor_port_number); ++ return; + } + } + /* if all aggregator's ports are READY_N == TRUE, set ready=TRUE +-- +2.35.1 + diff --git a/queue-5.4/drm-i915-unpin-on-error-in-intel_vgpu_shadow_mm_pin.patch b/queue-5.4/drm-i915-unpin-on-error-in-intel_vgpu_shadow_mm_pin.patch new file mode 100644 index 00000000000..9b1f8c4ceda --- /dev/null +++ b/queue-5.4/drm-i915-unpin-on-error-in-intel_vgpu_shadow_mm_pin.patch @@ -0,0 +1,36 @@ +From ac32335c8c7724003ad254073e7dedbeee1cd13a Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 15 Nov 2022 16:15:18 +0300 +Subject: drm/i915: unpin on error in intel_vgpu_shadow_mm_pin() + +From: Dan Carpenter + +[ Upstream commit 3792fc508c095abd84b10ceae12bd773e61fdc36 ] + +Call intel_vgpu_unpin_mm() on this error path. + +Fixes: 418741480809 ("drm/i915/gvt: Adding ppgtt to GVT GEM context after shadow pdps settled.") +Signed-off-by: Dan Carpenter +Signed-off-by: Zhenyu Wang +Link: http://patchwork.freedesktop.org/patch/msgid/Y3OQ5tgZIVxyQ/WV@kili +Reviewed-by: Zhenyu Wang +Signed-off-by: Sasha Levin +--- + drivers/gpu/drm/i915/gvt/scheduler.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c +index 058dcd541644..c1dc225d8436 100644 +--- a/drivers/gpu/drm/i915/gvt/scheduler.c ++++ b/drivers/gpu/drm/i915/gvt/scheduler.c +@@ -632,6 +632,7 @@ static int prepare_workload(struct intel_vgpu_workload *workload) + + if (workload->shadow_mm->type != INTEL_GVT_MM_PPGTT || + !workload->shadow_mm->ppgtt_mm.shadowed) { ++ intel_vgpu_unpin_mm(workload->shadow_mm); + gvt_vgpu_err("workload shadow ppgtt isn't ready\n"); + return -EINVAL; + } +-- +2.35.1 + diff --git a/queue-5.4/ext4-correct-inconsistent-error-msg-in-nojournal-mod.patch b/queue-5.4/ext4-correct-inconsistent-error-msg-in-nojournal-mod.patch new file mode 100644 index 00000000000..7140d1558a2 --- /dev/null +++ b/queue-5.4/ext4-correct-inconsistent-error-msg-in-nojournal-mod.patch @@ -0,0 +1,55 @@ +From 04c215fe03ab45d3fbd55d6d4cb077f0828b777e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 9 Nov 2022 15:43:43 +0800 +Subject: ext4: correct inconsistent error msg in nojournal mode + +From: Baokun Li + +[ Upstream commit 89481b5fa8c0640e62ba84c6020cee895f7ac643 ] + +When we used the journal_async_commit mounting option in nojournal mode, +the kernel told me that "can't mount with journal_checksum", was very +confusing. I find that when we mount with journal_async_commit, both the +JOURNAL_ASYNC_COMMIT and EXPLICIT_JOURNAL_CHECKSUM flags are set. However, +in the error branch, CHECKSUM is checked before ASYNC_COMMIT. As a result, +the above inconsistency occurs, and the ASYNC_COMMIT branch becomes dead +code that cannot be executed. Therefore, we exchange the positions of the +two judgments to make the error msg more accurate. + +Signed-off-by: Baokun Li +Reviewed-by: Jan Kara +Link: https://lore.kernel.org/r/20221109074343.4184862-1-libaokun1@huawei.com +Signed-off-by: Theodore Ts'o +Cc: stable@kernel.org +Signed-off-by: Sasha Levin +--- + fs/ext4/super.c | 9 +++++---- + 1 file changed, 5 insertions(+), 4 deletions(-) + +diff --git a/fs/ext4/super.c b/fs/ext4/super.c +index e3e616c55d0d..932c1619cea5 100644 +--- a/fs/ext4/super.c ++++ b/fs/ext4/super.c +@@ -4391,14 +4391,15 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) + goto failed_mount3a; + } else { + /* Nojournal mode, all journal mount options are illegal */ +- if (test_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM)) { ++ if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) { + ext4_msg(sb, KERN_ERR, "can't mount with " +- "journal_checksum, fs mounted w/o journal"); ++ "journal_async_commit, fs mounted w/o journal"); + goto failed_mount3a; + } +- if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) { ++ ++ if (test_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM)) { + ext4_msg(sb, KERN_ERR, "can't mount with " +- "journal_async_commit, fs mounted w/o journal"); ++ "journal_checksum, fs mounted w/o journal"); + goto failed_mount3a; + } + if (sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ) { +-- +2.35.1 + diff --git a/queue-5.4/ext4-fix-deadlock-due-to-mbcache-entry-corruption.patch b/queue-5.4/ext4-fix-deadlock-due-to-mbcache-entry-corruption.patch new file mode 100644 index 00000000000..11d0ce4c603 --- /dev/null +++ b/queue-5.4/ext4-fix-deadlock-due-to-mbcache-entry-corruption.patch @@ -0,0 +1,143 @@ +From 52350ff80a6fee8019734d107d455f9a4abefec2 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 23 Nov 2022 20:39:50 +0100 +Subject: ext4: fix deadlock due to mbcache entry corruption + +From: Jan Kara + +[ Upstream commit a44e84a9b7764c72896f7241a0ec9ac7e7ef38dd ] + +When manipulating xattr blocks, we can deadlock infinitely looping +inside ext4_xattr_block_set() where we constantly keep finding xattr +block for reuse in mbcache but we are unable to reuse it because its +reference count is too big. This happens because cache entry for the +xattr block is marked as reusable (e_reusable set) although its +reference count is too big. When this inconsistency happens, this +inconsistent state is kept indefinitely and so ext4_xattr_block_set() +keeps retrying indefinitely. + +The inconsistent state is caused by non-atomic update of e_reusable bit. +e_reusable is part of a bitfield and e_reusable update can race with +update of e_referenced bit in the same bitfield resulting in loss of one +of the updates. Fix the problem by using atomic bitops instead. + +This bug has been around for many years, but it became *much* easier +to hit after commit 65f8b80053a1 ("ext4: fix race when reusing xattr +blocks"). + +Cc: stable@vger.kernel.org +Fixes: 6048c64b2609 ("mbcache: add reusable flag to cache entries") +Fixes: 65f8b80053a1 ("ext4: fix race when reusing xattr blocks") +Reported-and-tested-by: Jeremi Piotrowski +Reported-by: Thilo Fromm +Link: https://lore.kernel.org/r/c77bf00f-4618-7149-56f1-b8d1664b9d07@linux.microsoft.com/ +Signed-off-by: Jan Kara +Reviewed-by: Andreas Dilger +Link: https://lore.kernel.org/r/20221123193950.16758-1-jack@suse.cz +Signed-off-by: Theodore Ts'o +Signed-off-by: Sasha Levin +--- + fs/ext4/xattr.c | 4 ++-- + fs/mbcache.c | 14 ++++++++------ + include/linux/mbcache.h | 9 +++++++-- + 3 files changed, 17 insertions(+), 10 deletions(-) + +diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c +index 131de3fcd2be..78df2d65998e 100644 +--- a/fs/ext4/xattr.c ++++ b/fs/ext4/xattr.c +@@ -1293,7 +1293,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode, + ce = mb_cache_entry_get(ea_block_cache, hash, + bh->b_blocknr); + if (ce) { +- ce->e_reusable = 1; ++ set_bit(MBE_REUSABLE_B, &ce->e_flags); + mb_cache_entry_put(ea_block_cache, ce); + } + } +@@ -2054,7 +2054,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + } + BHDR(new_bh)->h_refcount = cpu_to_le32(ref); + if (ref == EXT4_XATTR_REFCOUNT_MAX) +- ce->e_reusable = 0; ++ clear_bit(MBE_REUSABLE_B, &ce->e_flags); + ea_bdebug(new_bh, "reusing; refcount now=%d", + ref); + ext4_xattr_block_csum_set(inode, new_bh); +diff --git a/fs/mbcache.c b/fs/mbcache.c +index 950f1829a7fd..7a12ae87c806 100644 +--- a/fs/mbcache.c ++++ b/fs/mbcache.c +@@ -94,8 +94,9 @@ int mb_cache_entry_create(struct mb_cache *cache, gfp_t mask, u32 key, + atomic_set(&entry->e_refcnt, 1); + entry->e_key = key; + entry->e_value = value; +- entry->e_reusable = reusable; +- entry->e_referenced = 0; ++ entry->e_flags = 0; ++ if (reusable) ++ set_bit(MBE_REUSABLE_B, &entry->e_flags); + head = mb_cache_entry_head(cache, key); + hlist_bl_lock(head); + hlist_bl_for_each_entry(dup, dup_node, head, e_hash_list) { +@@ -162,7 +163,8 @@ static struct mb_cache_entry *__entry_find(struct mb_cache *cache, + while (node) { + entry = hlist_bl_entry(node, struct mb_cache_entry, + e_hash_list); +- if (entry->e_key == key && entry->e_reusable && ++ if (entry->e_key == key && ++ test_bit(MBE_REUSABLE_B, &entry->e_flags) && + atomic_inc_not_zero(&entry->e_refcnt)) + goto out; + node = node->next; +@@ -318,7 +320,7 @@ EXPORT_SYMBOL(mb_cache_entry_delete_or_get); + void mb_cache_entry_touch(struct mb_cache *cache, + struct mb_cache_entry *entry) + { +- entry->e_referenced = 1; ++ set_bit(MBE_REFERENCED_B, &entry->e_flags); + } + EXPORT_SYMBOL(mb_cache_entry_touch); + +@@ -343,9 +345,9 @@ static unsigned long mb_cache_shrink(struct mb_cache *cache, + entry = list_first_entry(&cache->c_list, + struct mb_cache_entry, e_list); + /* Drop initial hash reference if there is no user */ +- if (entry->e_referenced || ++ if (test_bit(MBE_REFERENCED_B, &entry->e_flags) || + atomic_cmpxchg(&entry->e_refcnt, 1, 0) != 1) { +- entry->e_referenced = 0; ++ clear_bit(MBE_REFERENCED_B, &entry->e_flags); + list_move_tail(&entry->e_list, &cache->c_list); + continue; + } +diff --git a/include/linux/mbcache.h b/include/linux/mbcache.h +index e9d5ece87794..591bc4cefe1d 100644 +--- a/include/linux/mbcache.h ++++ b/include/linux/mbcache.h +@@ -10,6 +10,12 @@ + + struct mb_cache; + ++/* Cache entry flags */ ++enum { ++ MBE_REFERENCED_B = 0, ++ MBE_REUSABLE_B ++}; ++ + struct mb_cache_entry { + /* List of entries in cache - protected by cache->c_list_lock */ + struct list_head e_list; +@@ -26,8 +32,7 @@ struct mb_cache_entry { + atomic_t e_refcnt; + /* Key in hash - stable during lifetime of the entry */ + u32 e_key; +- u32 e_referenced:1; +- u32 e_reusable:1; ++ unsigned long e_flags; + /* User provided value - stable during lifetime of the entry */ + u64 e_value; + }; +-- +2.35.1 + diff --git a/queue-5.4/ext4-fix-race-when-reusing-xattr-blocks.patch b/queue-5.4/ext4-fix-race-when-reusing-xattr-blocks.patch new file mode 100644 index 00000000000..dd218fe1917 --- /dev/null +++ b/queue-5.4/ext4-fix-race-when-reusing-xattr-blocks.patch @@ -0,0 +1,180 @@ +From 1c924039e9d8cb8f9964dd8a5f5c3a811bd37cab Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 12 Jul 2022 12:54:24 +0200 +Subject: ext4: fix race when reusing xattr blocks + +From: Jan Kara + +[ Upstream commit 65f8b80053a1b2fd602daa6814e62d6fa90e5e9b ] + +When ext4_xattr_block_set() decides to remove xattr block the following +race can happen: + +CPU1 CPU2 +ext4_xattr_block_set() ext4_xattr_release_block() + new_bh = ext4_xattr_block_cache_find() + + lock_buffer(bh); + ref = le32_to_cpu(BHDR(bh)->h_refcount); + if (ref == 1) { + ... + mb_cache_entry_delete(); + unlock_buffer(bh); + ext4_free_blocks(); + ... + ext4_forget(..., bh, ...); + jbd2_journal_revoke(..., bh); + + ext4_journal_get_write_access(..., new_bh, ...) + do_get_write_access() + jbd2_journal_cancel_revoke(..., new_bh); + +Later the code in ext4_xattr_block_set() finds out the block got freed +and cancels reusal of the block but the revoke stays canceled and so in +case of block reuse and journal replay the filesystem can get corrupted. +If the race works out slightly differently, we can also hit assertions +in the jbd2 code. + +Fix the problem by making sure that once matching mbcache entry is +found, code dropping the last xattr block reference (or trying to modify +xattr block in place) waits until the mbcache entry reference is +dropped. This way code trying to reuse xattr block is protected from +someone trying to drop the last reference to xattr block. + +Reported-and-tested-by: Ritesh Harjani +CC: stable@vger.kernel.org +Fixes: 82939d7999df ("ext4: convert to mbcache2") +Signed-off-by: Jan Kara +Link: https://lore.kernel.org/r/20220712105436.32204-5-jack@suse.cz +Signed-off-by: Theodore Ts'o +Stable-dep-of: a44e84a9b776 ("ext4: fix deadlock due to mbcache entry corruption") +Signed-off-by: Sasha Levin +--- + fs/ext4/xattr.c | 67 +++++++++++++++++++++++++++++++++---------------- + 1 file changed, 45 insertions(+), 22 deletions(-) + +diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c +index 24a4396933c0..131de3fcd2be 100644 +--- a/fs/ext4/xattr.c ++++ b/fs/ext4/xattr.c +@@ -437,9 +437,16 @@ static int ext4_xattr_inode_iget(struct inode *parent, unsigned long ea_ino, + /* Remove entry from mbcache when EA inode is getting evicted */ + void ext4_evict_ea_inode(struct inode *inode) + { +- if (EA_INODE_CACHE(inode)) +- mb_cache_entry_delete(EA_INODE_CACHE(inode), +- ext4_xattr_inode_get_hash(inode), inode->i_ino); ++ struct mb_cache_entry *oe; ++ ++ if (!EA_INODE_CACHE(inode)) ++ return; ++ /* Wait for entry to get unused so that we can remove it */ ++ while ((oe = mb_cache_entry_delete_or_get(EA_INODE_CACHE(inode), ++ ext4_xattr_inode_get_hash(inode), inode->i_ino))) { ++ mb_cache_entry_wait_unused(oe); ++ mb_cache_entry_put(EA_INODE_CACHE(inode), oe); ++ } + } + + static int +@@ -1241,6 +1248,7 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode, + if (error) + goto out; + ++retry_ref: + lock_buffer(bh); + hash = le32_to_cpu(BHDR(bh)->h_hash); + ref = le32_to_cpu(BHDR(bh)->h_refcount); +@@ -1250,9 +1258,18 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode, + * This must happen under buffer lock for + * ext4_xattr_block_set() to reliably detect freed block + */ +- if (ea_block_cache) +- mb_cache_entry_delete(ea_block_cache, hash, +- bh->b_blocknr); ++ if (ea_block_cache) { ++ struct mb_cache_entry *oe; ++ ++ oe = mb_cache_entry_delete_or_get(ea_block_cache, hash, ++ bh->b_blocknr); ++ if (oe) { ++ unlock_buffer(bh); ++ mb_cache_entry_wait_unused(oe); ++ mb_cache_entry_put(ea_block_cache, oe); ++ goto retry_ref; ++ } ++ } + get_bh(bh); + unlock_buffer(bh); + +@@ -1879,9 +1896,20 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + * ext4_xattr_block_set() to reliably detect modified + * block + */ +- if (ea_block_cache) +- mb_cache_entry_delete(ea_block_cache, hash, +- bs->bh->b_blocknr); ++ if (ea_block_cache) { ++ struct mb_cache_entry *oe; ++ ++ oe = mb_cache_entry_delete_or_get(ea_block_cache, ++ hash, bs->bh->b_blocknr); ++ if (oe) { ++ /* ++ * Xattr block is getting reused. Leave ++ * it alone. ++ */ ++ mb_cache_entry_put(ea_block_cache, oe); ++ goto clone_block; ++ } ++ } + ea_bdebug(bs->bh, "modifying in-place"); + error = ext4_xattr_set_entry(i, s, handle, inode, + true /* is_block */); +@@ -1897,6 +1925,7 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + goto cleanup; + goto inserted; + } ++clone_block: + unlock_buffer(bs->bh); + ea_bdebug(bs->bh, "cloning"); + s->base = kmemdup(BHDR(bs->bh), bs->bh->b_size, GFP_NOFS); +@@ -2002,18 +2031,13 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + lock_buffer(new_bh); + /* + * We have to be careful about races with +- * freeing, rehashing or adding references to +- * xattr block. Once we hold buffer lock xattr +- * block's state is stable so we can check +- * whether the block got freed / rehashed or +- * not. Since we unhash mbcache entry under +- * buffer lock when freeing / rehashing xattr +- * block, checking whether entry is still +- * hashed is reliable. Same rules hold for +- * e_reusable handling. ++ * adding references to xattr block. Once we ++ * hold buffer lock xattr block's state is ++ * stable so we can check the additional ++ * reference fits. + */ +- if (hlist_bl_unhashed(&ce->e_hash_list) || +- !ce->e_reusable) { ++ ref = le32_to_cpu(BHDR(new_bh)->h_refcount) + 1; ++ if (ref > EXT4_XATTR_REFCOUNT_MAX) { + /* + * Undo everything and check mbcache + * again. +@@ -2028,9 +2052,8 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + new_bh = NULL; + goto inserted; + } +- ref = le32_to_cpu(BHDR(new_bh)->h_refcount) + 1; + BHDR(new_bh)->h_refcount = cpu_to_le32(ref); +- if (ref >= EXT4_XATTR_REFCOUNT_MAX) ++ if (ref == EXT4_XATTR_REFCOUNT_MAX) + ce->e_reusable = 0; + ea_bdebug(new_bh, "reusing; refcount now=%d", + ref); +-- +2.35.1 + diff --git a/queue-5.4/ext4-goto-right-label-failed_mount3a.patch b/queue-5.4/ext4-goto-right-label-failed_mount3a.patch new file mode 100644 index 00000000000..bd6f27aeadc --- /dev/null +++ b/queue-5.4/ext4-goto-right-label-failed_mount3a.patch @@ -0,0 +1,69 @@ +From 85bcc3077f205090a967f47e87165324a172359d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 16 Sep 2022 22:15:12 +0800 +Subject: ext4: goto right label 'failed_mount3a' + +From: Jason Yan + +[ Upstream commit 43bd6f1b49b61f43de4d4e33661b8dbe8c911f14 ] + +Before these two branches neither loaded the journal nor created the +xattr cache. So the right label to goto is 'failed_mount3a'. Although +this did not cause any issues because the error handler validated if the +pointer is null. However this still made me confused when reading +the code. So it's still worth to modify to goto the right label. + +Signed-off-by: Jason Yan +Reviewed-by: Jan Kara +Reviewed-by: Ritesh Harjani (IBM) +Link: https://lore.kernel.org/r/20220916141527.1012715-2-yanaijie@huawei.com +Signed-off-by: Theodore Ts'o +Stable-dep-of: 89481b5fa8c0 ("ext4: correct inconsistent error msg in nojournal mode") +Signed-off-by: Sasha Levin +--- + fs/ext4/super.c | 10 +++++----- + 1 file changed, 5 insertions(+), 5 deletions(-) + +diff --git a/fs/ext4/super.c b/fs/ext4/super.c +index 5fdf584101e6..e3e616c55d0d 100644 +--- a/fs/ext4/super.c ++++ b/fs/ext4/super.c +@@ -4388,30 +4388,30 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) + ext4_has_feature_journal_needs_recovery(sb)) { + ext4_msg(sb, KERN_ERR, "required journal recovery " + "suppressed and not mounted read-only"); +- goto failed_mount_wq; ++ goto failed_mount3a; + } else { + /* Nojournal mode, all journal mount options are illegal */ + if (test_opt2(sb, EXPLICIT_JOURNAL_CHECKSUM)) { + ext4_msg(sb, KERN_ERR, "can't mount with " + "journal_checksum, fs mounted w/o journal"); +- goto failed_mount_wq; ++ goto failed_mount3a; + } + if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) { + ext4_msg(sb, KERN_ERR, "can't mount with " + "journal_async_commit, fs mounted w/o journal"); +- goto failed_mount_wq; ++ goto failed_mount3a; + } + if (sbi->s_commit_interval != JBD2_DEFAULT_MAX_COMMIT_AGE*HZ) { + ext4_msg(sb, KERN_ERR, "can't mount with " + "commit=%lu, fs mounted w/o journal", + sbi->s_commit_interval / HZ); +- goto failed_mount_wq; ++ goto failed_mount3a; + } + if (EXT4_MOUNT_DATA_FLAGS & + (sbi->s_mount_opt ^ sbi->s_def_mount_opt)) { + ext4_msg(sb, KERN_ERR, "can't mount with " + "data=, fs mounted w/o journal"); +- goto failed_mount_wq; ++ goto failed_mount3a; + } + sbi->s_def_mount_opt &= ~EXT4_MOUNT_JOURNAL_CHECKSUM; + clear_opt(sb, JOURNAL_CHECKSUM); +-- +2.35.1 + diff --git a/queue-5.4/ext4-remove-ea-inode-entry-from-mbcache-on-inode-evi.patch b/queue-5.4/ext4-remove-ea-inode-entry-from-mbcache-on-inode-evi.patch new file mode 100644 index 00000000000..8b7d04fe445 --- /dev/null +++ b/queue-5.4/ext4-remove-ea-inode-entry-from-mbcache-on-inode-evi.patch @@ -0,0 +1,117 @@ +From 9fe868e42d4ee415f73237f1bc063c2467deb6f9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 12 Jul 2022 12:54:22 +0200 +Subject: ext4: remove EA inode entry from mbcache on inode eviction + +From: Jan Kara + +[ Upstream commit 6bc0d63dad7f9f54d381925ee855b402f652fa39 ] + +Currently we remove EA inode from mbcache as soon as its xattr refcount +drops to zero. However there can be pending attempts to reuse the inode +and thus refcount handling code has to handle the situation when +refcount increases from zero anyway. So save some work and just keep EA +inode in mbcache until it is getting evicted. At that moment we are sure +following iget() of EA inode will fail anyway (or wait for eviction to +finish and load things from the disk again) and so removing mbcache +entry at that moment is fine and simplifies the code a bit. + +CC: stable@vger.kernel.org +Fixes: 82939d7999df ("ext4: convert to mbcache2") +Signed-off-by: Jan Kara +Link: https://lore.kernel.org/r/20220712105436.32204-3-jack@suse.cz +Signed-off-by: Theodore Ts'o +Stable-dep-of: a44e84a9b776 ("ext4: fix deadlock due to mbcache entry corruption") +Signed-off-by: Sasha Levin +--- + fs/ext4/inode.c | 2 ++ + fs/ext4/xattr.c | 24 ++++++++---------------- + fs/ext4/xattr.h | 1 + + 3 files changed, 11 insertions(+), 16 deletions(-) + +diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c +index a26e5ed6d61c..b38427b8d083 100644 +--- a/fs/ext4/inode.c ++++ b/fs/ext4/inode.c +@@ -207,6 +207,8 @@ void ext4_evict_inode(struct inode *inode) + + trace_ext4_evict_inode(inode); + ++ if (EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL) ++ ext4_evict_ea_inode(inode); + if (inode->i_nlink) { + /* + * When journalling data dirty buffers are tracked only in the +diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c +index 8f0e8b60ea20..4ade87a32315 100644 +--- a/fs/ext4/xattr.c ++++ b/fs/ext4/xattr.c +@@ -434,6 +434,14 @@ static int ext4_xattr_inode_iget(struct inode *parent, unsigned long ea_ino, + return err; + } + ++/* Remove entry from mbcache when EA inode is getting evicted */ ++void ext4_evict_ea_inode(struct inode *inode) ++{ ++ if (EA_INODE_CACHE(inode)) ++ mb_cache_entry_delete(EA_INODE_CACHE(inode), ++ ext4_xattr_inode_get_hash(inode), inode->i_ino); ++} ++ + static int + ext4_xattr_inode_verify_hashes(struct inode *ea_inode, + struct ext4_xattr_entry *entry, void *buffer, +@@ -1019,10 +1027,8 @@ static int ext4_xattr_ensure_credits(handle_t *handle, struct inode *inode, + static int ext4_xattr_inode_update_ref(handle_t *handle, struct inode *ea_inode, + int ref_change) + { +- struct mb_cache *ea_inode_cache = EA_INODE_CACHE(ea_inode); + struct ext4_iloc iloc; + s64 ref_count; +- u32 hash; + int ret; + + inode_lock(ea_inode); +@@ -1045,14 +1051,6 @@ static int ext4_xattr_inode_update_ref(handle_t *handle, struct inode *ea_inode, + + set_nlink(ea_inode, 1); + ext4_orphan_del(handle, ea_inode); +- +- if (ea_inode_cache) { +- hash = ext4_xattr_inode_get_hash(ea_inode); +- mb_cache_entry_create(ea_inode_cache, +- GFP_NOFS, hash, +- ea_inode->i_ino, +- true /* reusable */); +- } + } + } else { + WARN_ONCE(ref_count < 0, "EA inode %lu ref_count=%lld", +@@ -1065,12 +1063,6 @@ static int ext4_xattr_inode_update_ref(handle_t *handle, struct inode *ea_inode, + + clear_nlink(ea_inode); + ext4_orphan_add(handle, ea_inode); +- +- if (ea_inode_cache) { +- hash = ext4_xattr_inode_get_hash(ea_inode); +- mb_cache_entry_delete(ea_inode_cache, hash, +- ea_inode->i_ino); +- } + } + } + +diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h +index 990084e00374..231ef308d10c 100644 +--- a/fs/ext4/xattr.h ++++ b/fs/ext4/xattr.h +@@ -190,6 +190,7 @@ extern void ext4_xattr_inode_array_free(struct ext4_xattr_inode_array *array); + + extern int ext4_expand_extra_isize_ea(struct inode *inode, int new_extra_isize, + struct ext4_inode *raw_inode, handle_t *handle); ++extern void ext4_evict_ea_inode(struct inode *inode); + + extern const struct xattr_handler *ext4_xattr_handlers[]; + +-- +2.35.1 + diff --git a/queue-5.4/ext4-unindent-codeblock-in-ext4_xattr_block_set.patch b/queue-5.4/ext4-unindent-codeblock-in-ext4_xattr_block_set.patch new file mode 100644 index 00000000000..b4abad01ff3 --- /dev/null +++ b/queue-5.4/ext4-unindent-codeblock-in-ext4_xattr_block_set.patch @@ -0,0 +1,126 @@ +From 263e64afed205ce0446d5281716cf8d4b564408c Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 12 Jul 2022 12:54:23 +0200 +Subject: ext4: unindent codeblock in ext4_xattr_block_set() + +From: Jan Kara + +[ Upstream commit fd48e9acdf26d0cbd80051de07d4a735d05d29b2 ] + +Remove unnecessary else (and thus indentation level) from a code block +in ext4_xattr_block_set(). It will also make following code changes +easier. No functional changes. + +CC: stable@vger.kernel.org +Fixes: 82939d7999df ("ext4: convert to mbcache2") +Signed-off-by: Jan Kara +Link: https://lore.kernel.org/r/20220712105436.32204-4-jack@suse.cz +Signed-off-by: Theodore Ts'o +Stable-dep-of: a44e84a9b776 ("ext4: fix deadlock due to mbcache entry corruption") +Signed-off-by: Sasha Levin +--- + fs/ext4/xattr.c | 77 ++++++++++++++++++++++++------------------------- + 1 file changed, 38 insertions(+), 39 deletions(-) + +diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c +index 4ade87a32315..24a4396933c0 100644 +--- a/fs/ext4/xattr.c ++++ b/fs/ext4/xattr.c +@@ -1863,6 +1863,8 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + #define header(x) ((struct ext4_xattr_header *)(x)) + + if (s->base) { ++ int offset = (char *)s->here - bs->bh->b_data; ++ + BUFFER_TRACE(bs->bh, "get_write_access"); + error = ext4_journal_get_write_access(handle, bs->bh); + if (error) +@@ -1894,49 +1896,46 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + if (error) + goto cleanup; + goto inserted; +- } else { +- int offset = (char *)s->here - bs->bh->b_data; ++ } ++ unlock_buffer(bs->bh); ++ ea_bdebug(bs->bh, "cloning"); ++ s->base = kmemdup(BHDR(bs->bh), bs->bh->b_size, GFP_NOFS); ++ error = -ENOMEM; ++ if (s->base == NULL) ++ goto cleanup; ++ s->first = ENTRY(header(s->base)+1); ++ header(s->base)->h_refcount = cpu_to_le32(1); ++ s->here = ENTRY(s->base + offset); ++ s->end = s->base + bs->bh->b_size; + +- unlock_buffer(bs->bh); +- ea_bdebug(bs->bh, "cloning"); +- s->base = kmemdup(BHDR(bs->bh), bs->bh->b_size, GFP_NOFS); +- error = -ENOMEM; +- if (s->base == NULL) ++ /* ++ * If existing entry points to an xattr inode, we need ++ * to prevent ext4_xattr_set_entry() from decrementing ++ * ref count on it because the reference belongs to the ++ * original block. In this case, make the entry look ++ * like it has an empty value. ++ */ ++ if (!s->not_found && s->here->e_value_inum) { ++ ea_ino = le32_to_cpu(s->here->e_value_inum); ++ error = ext4_xattr_inode_iget(inode, ea_ino, ++ le32_to_cpu(s->here->e_hash), ++ &tmp_inode); ++ if (error) + goto cleanup; +- s->first = ENTRY(header(s->base)+1); +- header(s->base)->h_refcount = cpu_to_le32(1); +- s->here = ENTRY(s->base + offset); +- s->end = s->base + bs->bh->b_size; + +- /* +- * If existing entry points to an xattr inode, we need +- * to prevent ext4_xattr_set_entry() from decrementing +- * ref count on it because the reference belongs to the +- * original block. In this case, make the entry look +- * like it has an empty value. +- */ +- if (!s->not_found && s->here->e_value_inum) { +- ea_ino = le32_to_cpu(s->here->e_value_inum); +- error = ext4_xattr_inode_iget(inode, ea_ino, +- le32_to_cpu(s->here->e_hash), +- &tmp_inode); +- if (error) +- goto cleanup; +- +- if (!ext4_test_inode_state(tmp_inode, +- EXT4_STATE_LUSTRE_EA_INODE)) { +- /* +- * Defer quota free call for previous +- * inode until success is guaranteed. +- */ +- old_ea_inode_quota = le32_to_cpu( +- s->here->e_value_size); +- } +- iput(tmp_inode); +- +- s->here->e_value_inum = 0; +- s->here->e_value_size = 0; ++ if (!ext4_test_inode_state(tmp_inode, ++ EXT4_STATE_LUSTRE_EA_INODE)) { ++ /* ++ * Defer quota free call for previous ++ * inode until success is guaranteed. ++ */ ++ old_ea_inode_quota = le32_to_cpu( ++ s->here->e_value_size); + } ++ iput(tmp_inode); ++ ++ s->here->e_value_inum = 0; ++ s->here->e_value_size = 0; + } + } else { + /* Allocate a buffer where we construct the new block. */ +-- +2.35.1 + diff --git a/queue-5.4/ext4-use-kmemdup-to-replace-kmalloc-memcpy.patch b/queue-5.4/ext4-use-kmemdup-to-replace-kmalloc-memcpy.patch new file mode 100644 index 00000000000..59abe21797b --- /dev/null +++ b/queue-5.4/ext4-use-kmemdup-to-replace-kmalloc-memcpy.patch @@ -0,0 +1,41 @@ +From e4ead00e1f53862a44eabf274f6a31a135bf6169 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 25 May 2022 11:01:20 +0800 +Subject: ext4: use kmemdup() to replace kmalloc + memcpy + +From: Shuqi Zhang + +[ Upstream commit 4efd9f0d120c55b08852ee5605dbb02a77089a5d ] + +Replace kmalloc + memcpy with kmemdup() + +Signed-off-by: Shuqi Zhang +Reviewed-by: Ritesh Harjani +Link: https://lore.kernel.org/r/20220525030120.803330-1-zhangshuqi3@huawei.com +Signed-off-by: Theodore Ts'o +Stable-dep-of: a44e84a9b776 ("ext4: fix deadlock due to mbcache entry corruption") +Signed-off-by: Sasha Levin +--- + fs/ext4/xattr.c | 3 +-- + 1 file changed, 1 insertion(+), 2 deletions(-) + +diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c +index 0081eab74b20..8f0e8b60ea20 100644 +--- a/fs/ext4/xattr.c ++++ b/fs/ext4/xattr.c +@@ -1907,11 +1907,10 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, + + unlock_buffer(bs->bh); + ea_bdebug(bs->bh, "cloning"); +- s->base = kmalloc(bs->bh->b_size, GFP_NOFS); ++ s->base = kmemdup(BHDR(bs->bh), bs->bh->b_size, GFP_NOFS); + error = -ENOMEM; + if (s->base == NULL) + goto cleanup; +- memcpy(s->base, BHDR(bs->bh), bs->bh->b_size); + s->first = ENTRY(header(s->base)+1); + header(s->base)->h_refcount = cpu_to_le32(1); + s->here = ENTRY(s->base + offset); +-- +2.35.1 + diff --git a/queue-5.4/ext4-use-memcpy_to_page-in-pagecache_write.patch b/queue-5.4/ext4-use-memcpy_to_page-in-pagecache_write.patch new file mode 100644 index 00000000000..5d1e1154fd3 --- /dev/null +++ b/queue-5.4/ext4-use-memcpy_to_page-in-pagecache_write.patch @@ -0,0 +1,44 @@ +From 0b8b519467585e564104f6fdeb80aeba83ac0bc9 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sun, 7 Feb 2021 11:04:23 -0800 +Subject: ext4: use memcpy_to_page() in pagecache_write() + +From: Chaitanya Kulkarni + +[ Upstream commit bd256fda92efe97b692dc72e246d35fa724d42d8 ] + +Signed-off-by: Chaitanya Kulkarni +Link: https://lore.kernel.org/r/20210207190425.38107-7-chaitanya.kulkarni@wdc.com +Signed-off-by: Theodore Ts'o +Stable-dep-of: 956510c0c743 ("fs: ext4: initialize fsdata in pagecache_write()") +Signed-off-by: Sasha Levin +--- + fs/ext4/verity.c | 5 +---- + 1 file changed, 1 insertion(+), 4 deletions(-) + +diff --git a/fs/ext4/verity.c b/fs/ext4/verity.c +index 6a30e54c1128..0c67b7060eb4 100644 +--- a/fs/ext4/verity.c ++++ b/fs/ext4/verity.c +@@ -80,7 +80,6 @@ static int pagecache_write(struct inode *inode, const void *buf, size_t count, + PAGE_SIZE - offset_in_page(pos)); + struct page *page; + void *fsdata; +- void *addr; + int res; + + res = pagecache_write_begin(NULL, inode->i_mapping, pos, n, 0, +@@ -88,9 +87,7 @@ static int pagecache_write(struct inode *inode, const void *buf, size_t count, + if (res) + return res; + +- addr = kmap_atomic(page); +- memcpy(addr + offset_in_page(pos), buf, n); +- kunmap_atomic(addr); ++ memcpy_to_page(page, offset_in_page(pos), buf, n); + + res = pagecache_write_end(NULL, inode->i_mapping, pos, n, n, + page, fsdata); +-- +2.35.1 + diff --git a/queue-5.4/fs-ext4-initialize-fsdata-in-pagecache_write.patch b/queue-5.4/fs-ext4-initialize-fsdata-in-pagecache_write.patch new file mode 100644 index 00000000000..7d62d9e4d19 --- /dev/null +++ b/queue-5.4/fs-ext4-initialize-fsdata-in-pagecache_write.patch @@ -0,0 +1,43 @@ +From 3311e1aa8f8765434003bbf646bd4ddd16c9ead1 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 21 Nov 2022 12:21:30 +0100 +Subject: fs: ext4: initialize fsdata in pagecache_write() + +From: Alexander Potapenko + +[ Upstream commit 956510c0c7439e90b8103aaeaf4da92878c622f0 ] + +When aops->write_begin() does not initialize fsdata, KMSAN reports +an error passing the latter to aops->write_end(). + +Fix this by unconditionally initializing fsdata. + +Cc: Eric Biggers +Fixes: c93d8f885809 ("ext4: add basic fs-verity support") +Reported-by: syzbot+9767be679ef5016b6082@syzkaller.appspotmail.com +Signed-off-by: Alexander Potapenko +Reviewed-by: Eric Biggers +Link: https://lore.kernel.org/r/20221121112134.407362-1-glider@google.com +Signed-off-by: Theodore Ts'o +Cc: stable@kernel.org +Signed-off-by: Sasha Levin +--- + fs/ext4/verity.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/fs/ext4/verity.c b/fs/ext4/verity.c +index 0c67b7060eb4..9879ea046e5a 100644 +--- a/fs/ext4/verity.c ++++ b/fs/ext4/verity.c +@@ -79,7 +79,7 @@ static int pagecache_write(struct inode *inode, const void *buf, size_t count, + size_t n = min_t(size_t, count, + PAGE_SIZE - offset_in_page(pos)); + struct page *page; +- void *fsdata; ++ void *fsdata = NULL; + int res; + + res = pagecache_write_begin(NULL, inode->i_mapping, pos, n, 0, +-- +2.35.1 + diff --git a/queue-5.4/kvm-nvmx-properly-expose-enable_usr_wait_pause-contr.patch b/queue-5.4/kvm-nvmx-properly-expose-enable_usr_wait_pause-contr.patch new file mode 100644 index 00000000000..fa145f86568 --- /dev/null +++ b/queue-5.4/kvm-nvmx-properly-expose-enable_usr_wait_pause-contr.patch @@ -0,0 +1,65 @@ +From de47e33cc38f2766203fc065a3323b800c7d59ee Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 13 Dec 2022 06:23:03 +0000 +Subject: KVM: nVMX: Properly expose ENABLE_USR_WAIT_PAUSE control to L1 + +From: Sean Christopherson + +[ Upstream commit 31de69f4eea77b28a9724b3fa55aae104fc91fc7 ] + +Set ENABLE_USR_WAIT_PAUSE in KVM's supported VMX MSR configuration if the +feature is supported in hardware and enabled in KVM's base, non-nested +configuration, i.e. expose ENABLE_USR_WAIT_PAUSE to L1 if it's supported. +This fixes a bug where saving/restoring, i.e. migrating, a vCPU will fail +if WAITPKG (the associated CPUID feature) is enabled for the vCPU, and +obviously allows L1 to enable the feature for L2. + +KVM already effectively exposes ENABLE_USR_WAIT_PAUSE to L1 by stuffing +the allowed-1 control ina vCPU's virtual MSR_IA32_VMX_PROCBASED_CTLS2 when +updating secondary controls in response to KVM_SET_CPUID(2), but (a) that +depends on flawed code (KVM shouldn't touch VMX MSRs in response to CPUID +updates) and (b) runs afoul of vmx_restore_control_msr()'s restriction +that the guest value must be a strict subset of the supported host value. + +Although no past commit explicitly enabled nested support for WAITPKG, +doing so is safe and functionally correct from an architectural +perspective as no additional KVM support is needed to virtualize TPAUSE, +UMONITOR, and UMWAIT for L2 relative to L1, and KVM already forwards +VM-Exits to L1 as necessary (commit bf653b78f960, "KVM: vmx: Introduce +handle_unexpected_vmexit and handle WAITPKG vmexit"). + +Note, KVM always keeps the hosts MSR_IA32_UMWAIT_CONTROL resident in +hardware, i.e. always runs both L1 and L2 with the host's power management +settings for TPAUSE and UMWAIT. See commit bf09fb6cba4f ("KVM: VMX: Stop +context switching MSR_IA32_UMWAIT_CONTROL") for more details. + +Fixes: e69e72faa3a0 ("KVM: x86: Add support for user wait instructions") +Cc: stable@vger.kernel.org +Reported-by: Aaron Lewis +Reported-by: Yu Zhang +Signed-off-by: Sean Christopherson +Reviewed-by: Jim Mattson +Message-Id: <20221213062306.667649-2-seanjc@google.com> +Signed-off-by: Paolo Bonzini +Signed-off-by: Sasha Levin +--- + arch/x86/kvm/vmx/nested.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c +index 1dd693d18395..00f3336194a9 100644 +--- a/arch/x86/kvm/vmx/nested.c ++++ b/arch/x86/kvm/vmx/nested.c +@@ -5916,7 +5916,8 @@ void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, u32 ept_caps) + SECONDARY_EXEC_RDRAND_EXITING | + SECONDARY_EXEC_ENABLE_INVPCID | + SECONDARY_EXEC_RDSEED_EXITING | +- SECONDARY_EXEC_XSAVES; ++ SECONDARY_EXEC_XSAVES | ++ SECONDARY_EXEC_ENABLE_USR_WAIT_PAUSE; + + /* + * We can emulate "VMCS shadowing," even if the hardware +-- +2.35.1 + diff --git a/queue-5.4/kvm-retpolines-x86-eliminate-retpoline-from-vmx.c-ex.patch b/queue-5.4/kvm-retpolines-x86-eliminate-retpoline-from-vmx.c-ex.patch new file mode 100644 index 00000000000..0c9bda4727a --- /dev/null +++ b/queue-5.4/kvm-retpolines-x86-eliminate-retpoline-from-vmx.c-ex.patch @@ -0,0 +1,62 @@ +From 3c1597d7ec45ff271a75103e2981fc3b6c58a96d Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 4 Nov 2019 17:59:59 -0500 +Subject: KVM: retpolines: x86: eliminate retpoline from vmx.c exit handlers + +From: Andrea Arcangeli + +[ Upstream commit 4289d2728664fc1fb49cfc76a6a7d96d913b921f ] + +It's enough to check the exit value and issue a direct call to avoid +the retpoline for all the common vmexit reasons. + +Of course CONFIG_RETPOLINE already forbids gcc to use indirect jumps +while compiling all switch() statements, however switch() would still +allow the compiler to bisect the case value. It's more efficient to +prioritize the most frequent vmexits instead. + +The halt may be slow paths from the point of the guest, but not +necessarily so from the point of the host if the host runs at full CPU +capacity and no host CPU is ever left idle. + +Signed-off-by: Andrea Arcangeli +Signed-off-by: Paolo Bonzini +Stable-dep-of: 31de69f4eea7 ("KVM: nVMX: Properly expose ENABLE_USR_WAIT_PAUSE control to L1") +Signed-off-by: Sasha Levin +--- + arch/x86/kvm/vmx/vmx.c | 18 ++++++++++++++++-- + 1 file changed, 16 insertions(+), 2 deletions(-) + +diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c +index 0fae9b448ab9..668505c6abe9 100644 +--- a/arch/x86/kvm/vmx/vmx.c ++++ b/arch/x86/kvm/vmx/vmx.c +@@ -6015,9 +6015,23 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu) + } + + if (exit_reason < kvm_vmx_max_exit_handlers +- && kvm_vmx_exit_handlers[exit_reason]) ++ && kvm_vmx_exit_handlers[exit_reason]) { ++#ifdef CONFIG_RETPOLINE ++ if (exit_reason == EXIT_REASON_MSR_WRITE) ++ return kvm_emulate_wrmsr(vcpu); ++ else if (exit_reason == EXIT_REASON_PREEMPTION_TIMER) ++ return handle_preemption_timer(vcpu); ++ else if (exit_reason == EXIT_REASON_PENDING_INTERRUPT) ++ return handle_interrupt_window(vcpu); ++ else if (exit_reason == EXIT_REASON_EXTERNAL_INTERRUPT) ++ return handle_external_interrupt(vcpu); ++ else if (exit_reason == EXIT_REASON_HLT) ++ return kvm_emulate_halt(vcpu); ++ else if (exit_reason == EXIT_REASON_EPT_MISCONFIG) ++ return handle_ept_misconfig(vcpu); ++#endif + return kvm_vmx_exit_handlers[exit_reason](vcpu); +- else { ++ } else { + vcpu_unimpl(vcpu, "vmx: unexpected exit reason 0x%x\n", + exit_reason); + dump_vmcs(); +-- +2.35.1 + diff --git a/queue-5.4/kvm-vmx-fix-the-spelling-of-cpu_based_use_tsc_offset.patch b/queue-5.4/kvm-vmx-fix-the-spelling-of-cpu_based_use_tsc_offset.patch new file mode 100644 index 00000000000..102cbe789a4 --- /dev/null +++ b/queue-5.4/kvm-vmx-fix-the-spelling-of-cpu_based_use_tsc_offset.patch @@ -0,0 +1,136 @@ +From 8f9706aac0be8c39909b04db5082921bafb8b19e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 6 Dec 2019 16:45:26 +0800 +Subject: KVM: VMX: Fix the spelling of CPU_BASED_USE_TSC_OFFSETTING + +From: Xiaoyao Li + +[ Upstream commit 5e3d394fdd9e6b49cd8b28d85adff100a5bddc66 ] + +The mis-spelling is found by checkpatch.pl, so fix them. + +Signed-off-by: Xiaoyao Li +Signed-off-by: Paolo Bonzini +Stable-dep-of: 31de69f4eea7 ("KVM: nVMX: Properly expose ENABLE_USR_WAIT_PAUSE control to L1") +Signed-off-by: Sasha Levin +--- + arch/x86/include/asm/vmx.h | 2 +- + arch/x86/kvm/vmx/nested.c | 8 ++++---- + arch/x86/kvm/vmx/vmx.c | 6 +++--- + tools/testing/selftests/kvm/include/x86_64/vmx.h | 2 +- + tools/testing/selftests/kvm/x86_64/vmx_tsc_adjust_test.c | 2 +- + 5 files changed, 10 insertions(+), 10 deletions(-) + +diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h +index 06d4420508c5..d716fe938fc0 100644 +--- a/arch/x86/include/asm/vmx.h ++++ b/arch/x86/include/asm/vmx.h +@@ -20,7 +20,7 @@ + * Definitions of Primary Processor-Based VM-Execution Controls. + */ + #define CPU_BASED_INTR_WINDOW_EXITING 0x00000004 +-#define CPU_BASED_USE_TSC_OFFSETING 0x00000008 ++#define CPU_BASED_USE_TSC_OFFSETTING 0x00000008 + #define CPU_BASED_HLT_EXITING 0x00000080 + #define CPU_BASED_INVLPG_EXITING 0x00000200 + #define CPU_BASED_MWAIT_EXITING 0x00000400 +diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c +index dca2c78db5d0..1dd693d18395 100644 +--- a/arch/x86/kvm/vmx/nested.c ++++ b/arch/x86/kvm/vmx/nested.c +@@ -3090,7 +3090,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu, + } + + enter_guest_mode(vcpu); +- if (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETING) ++ if (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETTING) + vcpu->arch.tsc_offset += vmcs12->tsc_offset; + + if (prepare_vmcs02(vcpu, vmcs12, &exit_qual)) +@@ -3154,7 +3154,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu, + * 26.7 "VM-entry failures during or after loading guest state". + */ + vmentry_fail_vmexit_guest_mode: +- if (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETING) ++ if (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETTING) + vcpu->arch.tsc_offset -= vmcs12->tsc_offset; + leave_guest_mode(vcpu); + +@@ -4073,7 +4073,7 @@ void nested_vmx_vmexit(struct kvm_vcpu *vcpu, u32 exit_reason, + if (nested_cpu_has_preemption_timer(vmcs12)) + hrtimer_cancel(&to_vmx(vcpu)->nested.preemption_timer); + +- if (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETING) ++ if (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETTING) + vcpu->arch.tsc_offset -= vmcs12->tsc_offset; + + if (likely(!vmx->fail)) { +@@ -5870,7 +5870,7 @@ void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, u32 ept_caps) + CPU_BASED_ALWAYSON_WITHOUT_TRUE_MSR; + msrs->procbased_ctls_high &= + CPU_BASED_INTR_WINDOW_EXITING | +- CPU_BASED_NMI_WINDOW_EXITING | CPU_BASED_USE_TSC_OFFSETING | ++ CPU_BASED_NMI_WINDOW_EXITING | CPU_BASED_USE_TSC_OFFSETTING | + CPU_BASED_HLT_EXITING | CPU_BASED_INVLPG_EXITING | + CPU_BASED_MWAIT_EXITING | CPU_BASED_CR3_LOAD_EXITING | + CPU_BASED_CR3_STORE_EXITING | +diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c +index 470a8f9a0046..df77207d93b0 100644 +--- a/arch/x86/kvm/vmx/vmx.c ++++ b/arch/x86/kvm/vmx/vmx.c +@@ -1780,7 +1780,7 @@ static u64 vmx_read_l1_tsc_offset(struct kvm_vcpu *vcpu) + struct vmcs12 *vmcs12 = get_vmcs12(vcpu); + + if (is_guest_mode(vcpu) && +- (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETING)) ++ (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETTING)) + return vcpu->arch.tsc_offset - vmcs12->tsc_offset; + + return vcpu->arch.tsc_offset; +@@ -1798,7 +1798,7 @@ static u64 vmx_write_l1_tsc_offset(struct kvm_vcpu *vcpu, u64 offset) + * to the newly set TSC to get L2's TSC. + */ + if (is_guest_mode(vcpu) && +- (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETING)) ++ (vmcs12->cpu_based_vm_exec_control & CPU_BASED_USE_TSC_OFFSETTING)) + g_tsc_offset = vmcs12->tsc_offset; + + trace_kvm_write_tsc_offset(vcpu->vcpu_id, +@@ -2425,7 +2425,7 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf, + CPU_BASED_CR3_STORE_EXITING | + CPU_BASED_UNCOND_IO_EXITING | + CPU_BASED_MOV_DR_EXITING | +- CPU_BASED_USE_TSC_OFFSETING | ++ CPU_BASED_USE_TSC_OFFSETTING | + CPU_BASED_MWAIT_EXITING | + CPU_BASED_MONITOR_EXITING | + CPU_BASED_INVLPG_EXITING | +diff --git a/tools/testing/selftests/kvm/include/x86_64/vmx.h b/tools/testing/selftests/kvm/include/x86_64/vmx.h +index 7eb38451c359..3d27069b9ed9 100644 +--- a/tools/testing/selftests/kvm/include/x86_64/vmx.h ++++ b/tools/testing/selftests/kvm/include/x86_64/vmx.h +@@ -19,7 +19,7 @@ + * Definitions of Primary Processor-Based VM-Execution Controls. + */ + #define CPU_BASED_INTR_WINDOW_EXITING 0x00000004 +-#define CPU_BASED_USE_TSC_OFFSETING 0x00000008 ++#define CPU_BASED_USE_TSC_OFFSETTING 0x00000008 + #define CPU_BASED_HLT_EXITING 0x00000080 + #define CPU_BASED_INVLPG_EXITING 0x00000200 + #define CPU_BASED_MWAIT_EXITING 0x00000400 +diff --git a/tools/testing/selftests/kvm/x86_64/vmx_tsc_adjust_test.c b/tools/testing/selftests/kvm/x86_64/vmx_tsc_adjust_test.c +index 5590fd2bcf87..69e482a95c47 100644 +--- a/tools/testing/selftests/kvm/x86_64/vmx_tsc_adjust_test.c ++++ b/tools/testing/selftests/kvm/x86_64/vmx_tsc_adjust_test.c +@@ -98,7 +98,7 @@ static void l1_guest_code(struct vmx_pages *vmx_pages) + prepare_vmcs(vmx_pages, l2_guest_code, + &l2_guest_stack[L2_GUEST_STACK_SIZE]); + control = vmreadz(CPU_BASED_VM_EXEC_CONTROL); +- control |= CPU_BASED_USE_MSR_BITMAPS | CPU_BASED_USE_TSC_OFFSETING; ++ control |= CPU_BASED_USE_MSR_BITMAPS | CPU_BASED_USE_TSC_OFFSETTING; + vmwrite(CPU_BASED_VM_EXEC_CONTROL, control); + vmwrite(TSC_OFFSET, TSC_OFFSET_VALUE); + +-- +2.35.1 + diff --git a/queue-5.4/kvm-vmx-rename-interrupt_pending-to-interrupt_window.patch b/queue-5.4/kvm-vmx-rename-interrupt_pending-to-interrupt_window.patch new file mode 100644 index 00000000000..85c98c6bf1c --- /dev/null +++ b/queue-5.4/kvm-vmx-rename-interrupt_pending-to-interrupt_window.patch @@ -0,0 +1,207 @@ +From e916ed024c3ded636f11be23596b2f86bfcb8053 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 6 Dec 2019 16:45:24 +0800 +Subject: KVM: VMX: Rename INTERRUPT_PENDING to INTERRUPT_WINDOW + +From: Xiaoyao Li + +[ Upstream commit 9dadc2f918df26e64aa04794cdb4d8667c934f47 ] + +Rename interrupt-windown exiting related definitions to match the +latest Intel SDM. No functional changes. + +Signed-off-by: Xiaoyao Li +Signed-off-by: Paolo Bonzini +Stable-dep-of: 31de69f4eea7 ("KVM: nVMX: Properly expose ENABLE_USR_WAIT_PAUSE control to L1") +Signed-off-by: Sasha Levin +--- + arch/x86/include/asm/vmx.h | 2 +- + arch/x86/include/uapi/asm/vmx.h | 4 ++-- + arch/x86/kvm/vmx/nested.c | 12 ++++++------ + arch/x86/kvm/vmx/vmx.c | 10 +++++----- + tools/arch/x86/include/uapi/asm/vmx.h | 4 ++-- + tools/testing/selftests/kvm/include/x86_64/vmx.h | 4 ++-- + 6 files changed, 18 insertions(+), 18 deletions(-) + +diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h +index 1835767aa335..5acda8d9b9a7 100644 +--- a/arch/x86/include/asm/vmx.h ++++ b/arch/x86/include/asm/vmx.h +@@ -19,7 +19,7 @@ + /* + * Definitions of Primary Processor-Based VM-Execution Controls. + */ +-#define CPU_BASED_VIRTUAL_INTR_PENDING 0x00000004 ++#define CPU_BASED_INTR_WINDOW_EXITING 0x00000004 + #define CPU_BASED_USE_TSC_OFFSETING 0x00000008 + #define CPU_BASED_HLT_EXITING 0x00000080 + #define CPU_BASED_INVLPG_EXITING 0x00000200 +diff --git a/arch/x86/include/uapi/asm/vmx.h b/arch/x86/include/uapi/asm/vmx.h +index 3eb8411ab60e..e95b72ec19bc 100644 +--- a/arch/x86/include/uapi/asm/vmx.h ++++ b/arch/x86/include/uapi/asm/vmx.h +@@ -33,7 +33,7 @@ + #define EXIT_REASON_TRIPLE_FAULT 2 + #define EXIT_REASON_INIT_SIGNAL 3 + +-#define EXIT_REASON_PENDING_INTERRUPT 7 ++#define EXIT_REASON_INTERRUPT_WINDOW 7 + #define EXIT_REASON_NMI_WINDOW 8 + #define EXIT_REASON_TASK_SWITCH 9 + #define EXIT_REASON_CPUID 10 +@@ -94,7 +94,7 @@ + { EXIT_REASON_EXTERNAL_INTERRUPT, "EXTERNAL_INTERRUPT" }, \ + { EXIT_REASON_TRIPLE_FAULT, "TRIPLE_FAULT" }, \ + { EXIT_REASON_INIT_SIGNAL, "INIT_SIGNAL" }, \ +- { EXIT_REASON_PENDING_INTERRUPT, "PENDING_INTERRUPT" }, \ ++ { EXIT_REASON_INTERRUPT_WINDOW, "INTERRUPT_WINDOW" }, \ + { EXIT_REASON_NMI_WINDOW, "NMI_WINDOW" }, \ + { EXIT_REASON_TASK_SWITCH, "TASK_SWITCH" }, \ + { EXIT_REASON_CPUID, "CPUID" }, \ +diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c +index 6ab0410f1030..ee768f977a0a 100644 +--- a/arch/x86/kvm/vmx/nested.c ++++ b/arch/x86/kvm/vmx/nested.c +@@ -2073,7 +2073,7 @@ static void prepare_vmcs02_early(struct vcpu_vmx *vmx, struct vmcs12 *vmcs12) + * EXEC CONTROLS + */ + exec_control = vmx_exec_control(vmx); /* L0's desires */ +- exec_control &= ~CPU_BASED_VIRTUAL_INTR_PENDING; ++ exec_control &= ~CPU_BASED_INTR_WINDOW_EXITING; + exec_control &= ~CPU_BASED_VIRTUAL_NMI_PENDING; + exec_control &= ~CPU_BASED_TPR_SHADOW; + exec_control |= vmcs12->cpu_based_vm_exec_control; +@@ -3039,7 +3039,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu, + u32 exit_qual; + + evaluate_pending_interrupts = exec_controls_get(vmx) & +- (CPU_BASED_VIRTUAL_INTR_PENDING | CPU_BASED_VIRTUAL_NMI_PENDING); ++ (CPU_BASED_INTR_WINDOW_EXITING | CPU_BASED_VIRTUAL_NMI_PENDING); + if (likely(!evaluate_pending_interrupts) && kvm_vcpu_apicv_active(vcpu)) + evaluate_pending_interrupts |= vmx_has_apicv_interrupt(vcpu); + +@@ -3268,7 +3268,7 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) + if ((vmcs12->guest_activity_state == GUEST_ACTIVITY_HLT) && + !(vmcs12->vm_entry_intr_info_field & INTR_INFO_VALID_MASK) && + !(vmcs12->cpu_based_vm_exec_control & CPU_BASED_VIRTUAL_NMI_PENDING) && +- !((vmcs12->cpu_based_vm_exec_control & CPU_BASED_VIRTUAL_INTR_PENDING) && ++ !((vmcs12->cpu_based_vm_exec_control & CPU_BASED_INTR_WINDOW_EXITING) && + (vmcs12->guest_rflags & X86_EFLAGS_IF))) { + vmx->nested.nested_run_pending = 0; + return kvm_vcpu_halt(vcpu); +@@ -5376,8 +5376,8 @@ bool nested_vmx_exit_reflected(struct kvm_vcpu *vcpu, u32 exit_reason) + return false; + case EXIT_REASON_TRIPLE_FAULT: + return true; +- case EXIT_REASON_PENDING_INTERRUPT: +- return nested_cpu_has(vmcs12, CPU_BASED_VIRTUAL_INTR_PENDING); ++ case EXIT_REASON_INTERRUPT_WINDOW: ++ return nested_cpu_has(vmcs12, CPU_BASED_INTR_WINDOW_EXITING); + case EXIT_REASON_NMI_WINDOW: + return nested_cpu_has(vmcs12, CPU_BASED_VIRTUAL_NMI_PENDING); + case EXIT_REASON_TASK_SWITCH: +@@ -5869,7 +5869,7 @@ void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, u32 ept_caps) + msrs->procbased_ctls_low = + CPU_BASED_ALWAYSON_WITHOUT_TRUE_MSR; + msrs->procbased_ctls_high &= +- CPU_BASED_VIRTUAL_INTR_PENDING | ++ CPU_BASED_INTR_WINDOW_EXITING | + CPU_BASED_VIRTUAL_NMI_PENDING | CPU_BASED_USE_TSC_OFFSETING | + CPU_BASED_HLT_EXITING | CPU_BASED_INVLPG_EXITING | + CPU_BASED_MWAIT_EXITING | CPU_BASED_CR3_LOAD_EXITING | +diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c +index 668505c6abe9..51aa5851011c 100644 +--- a/arch/x86/kvm/vmx/vmx.c ++++ b/arch/x86/kvm/vmx/vmx.c +@@ -4458,7 +4458,7 @@ static void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) + + static void enable_irq_window(struct kvm_vcpu *vcpu) + { +- exec_controls_setbit(to_vmx(vcpu), CPU_BASED_VIRTUAL_INTR_PENDING); ++ exec_controls_setbit(to_vmx(vcpu), CPU_BASED_INTR_WINDOW_EXITING); + } + + static void enable_nmi_window(struct kvm_vcpu *vcpu) +@@ -5082,7 +5082,7 @@ static int handle_tpr_below_threshold(struct kvm_vcpu *vcpu) + + static int handle_interrupt_window(struct kvm_vcpu *vcpu) + { +- exec_controls_clearbit(to_vmx(vcpu), CPU_BASED_VIRTUAL_INTR_PENDING); ++ exec_controls_clearbit(to_vmx(vcpu), CPU_BASED_INTR_WINDOW_EXITING); + + kvm_make_request(KVM_REQ_EVENT, vcpu); + +@@ -5316,7 +5316,7 @@ static int handle_invalid_guest_state(struct kvm_vcpu *vcpu) + WARN_ON_ONCE(vmx->emulation_required && vmx->nested.nested_run_pending); + + intr_window_requested = exec_controls_get(vmx) & +- CPU_BASED_VIRTUAL_INTR_PENDING; ++ CPU_BASED_INTR_WINDOW_EXITING; + + while (vmx->emulation_required && count-- != 0) { + if (intr_window_requested && vmx_interrupt_allowed(vcpu)) +@@ -5640,7 +5640,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = { + [EXIT_REASON_CPUID] = kvm_emulate_cpuid, + [EXIT_REASON_MSR_READ] = kvm_emulate_rdmsr, + [EXIT_REASON_MSR_WRITE] = kvm_emulate_wrmsr, +- [EXIT_REASON_PENDING_INTERRUPT] = handle_interrupt_window, ++ [EXIT_REASON_INTERRUPT_WINDOW] = handle_interrupt_window, + [EXIT_REASON_HLT] = kvm_emulate_halt, + [EXIT_REASON_INVD] = handle_invd, + [EXIT_REASON_INVLPG] = handle_invlpg, +@@ -6021,7 +6021,7 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu) + return kvm_emulate_wrmsr(vcpu); + else if (exit_reason == EXIT_REASON_PREEMPTION_TIMER) + return handle_preemption_timer(vcpu); +- else if (exit_reason == EXIT_REASON_PENDING_INTERRUPT) ++ else if (exit_reason == EXIT_REASON_INTERRUPT_WINDOW) + return handle_interrupt_window(vcpu); + else if (exit_reason == EXIT_REASON_EXTERNAL_INTERRUPT) + return handle_external_interrupt(vcpu); +diff --git a/tools/arch/x86/include/uapi/asm/vmx.h b/tools/arch/x86/include/uapi/asm/vmx.h +index 3eb8411ab60e..e95b72ec19bc 100644 +--- a/tools/arch/x86/include/uapi/asm/vmx.h ++++ b/tools/arch/x86/include/uapi/asm/vmx.h +@@ -33,7 +33,7 @@ + #define EXIT_REASON_TRIPLE_FAULT 2 + #define EXIT_REASON_INIT_SIGNAL 3 + +-#define EXIT_REASON_PENDING_INTERRUPT 7 ++#define EXIT_REASON_INTERRUPT_WINDOW 7 + #define EXIT_REASON_NMI_WINDOW 8 + #define EXIT_REASON_TASK_SWITCH 9 + #define EXIT_REASON_CPUID 10 +@@ -94,7 +94,7 @@ + { EXIT_REASON_EXTERNAL_INTERRUPT, "EXTERNAL_INTERRUPT" }, \ + { EXIT_REASON_TRIPLE_FAULT, "TRIPLE_FAULT" }, \ + { EXIT_REASON_INIT_SIGNAL, "INIT_SIGNAL" }, \ +- { EXIT_REASON_PENDING_INTERRUPT, "PENDING_INTERRUPT" }, \ ++ { EXIT_REASON_INTERRUPT_WINDOW, "INTERRUPT_WINDOW" }, \ + { EXIT_REASON_NMI_WINDOW, "NMI_WINDOW" }, \ + { EXIT_REASON_TASK_SWITCH, "TASK_SWITCH" }, \ + { EXIT_REASON_CPUID, "CPUID" }, \ +diff --git a/tools/testing/selftests/kvm/include/x86_64/vmx.h b/tools/testing/selftests/kvm/include/x86_64/vmx.h +index f52e0ba84fed..c6e442d7a241 100644 +--- a/tools/testing/selftests/kvm/include/x86_64/vmx.h ++++ b/tools/testing/selftests/kvm/include/x86_64/vmx.h +@@ -18,7 +18,7 @@ + /* + * Definitions of Primary Processor-Based VM-Execution Controls. + */ +-#define CPU_BASED_VIRTUAL_INTR_PENDING 0x00000004 ++#define CPU_BASED_INTR_WINDOW_EXITING 0x00000004 + #define CPU_BASED_USE_TSC_OFFSETING 0x00000008 + #define CPU_BASED_HLT_EXITING 0x00000080 + #define CPU_BASED_INVLPG_EXITING 0x00000200 +@@ -103,7 +103,7 @@ + #define EXIT_REASON_EXCEPTION_NMI 0 + #define EXIT_REASON_EXTERNAL_INTERRUPT 1 + #define EXIT_REASON_TRIPLE_FAULT 2 +-#define EXIT_REASON_PENDING_INTERRUPT 7 ++#define EXIT_REASON_INTERRUPT_WINDOW 7 + #define EXIT_REASON_NMI_WINDOW 8 + #define EXIT_REASON_TASK_SWITCH 9 + #define EXIT_REASON_CPUID 10 +-- +2.35.1 + diff --git a/queue-5.4/kvm-vmx-rename-nmi_pending-to-nmi_window.patch b/queue-5.4/kvm-vmx-rename-nmi_pending-to-nmi_window.patch new file mode 100644 index 00000000000..ce02a04fd50 --- /dev/null +++ b/queue-5.4/kvm-vmx-rename-nmi_pending-to-nmi_window.patch @@ -0,0 +1,132 @@ +From 1049ab38cbf6c1232aac92176a3d4e63f690d93e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 6 Dec 2019 16:45:25 +0800 +Subject: KVM: VMX: Rename NMI_PENDING to NMI_WINDOW + +From: Xiaoyao Li + +[ Upstream commit 4e2a0bc56ad197e5ccfab8395649b681067fe8cb ] + +Rename the NMI-window exiting related definitions to match the latest +Intel SDM. No functional changes. + +Signed-off-by: Xiaoyao Li +Signed-off-by: Paolo Bonzini +Stable-dep-of: 31de69f4eea7 ("KVM: nVMX: Properly expose ENABLE_USR_WAIT_PAUSE control to L1") +Signed-off-by: Sasha Levin +--- + arch/x86/include/asm/vmx.h | 2 +- + arch/x86/kvm/vmx/nested.c | 12 ++++++------ + arch/x86/kvm/vmx/vmx.c | 4 ++-- + tools/testing/selftests/kvm/include/x86_64/vmx.h | 2 +- + 4 files changed, 10 insertions(+), 10 deletions(-) + +diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h +index 5acda8d9b9a7..06d4420508c5 100644 +--- a/arch/x86/include/asm/vmx.h ++++ b/arch/x86/include/asm/vmx.h +@@ -31,7 +31,7 @@ + #define CPU_BASED_CR8_LOAD_EXITING 0x00080000 + #define CPU_BASED_CR8_STORE_EXITING 0x00100000 + #define CPU_BASED_TPR_SHADOW 0x00200000 +-#define CPU_BASED_VIRTUAL_NMI_PENDING 0x00400000 ++#define CPU_BASED_NMI_WINDOW_EXITING 0x00400000 + #define CPU_BASED_MOV_DR_EXITING 0x00800000 + #define CPU_BASED_UNCOND_IO_EXITING 0x01000000 + #define CPU_BASED_USE_IO_BITMAPS 0x02000000 +diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c +index ee768f977a0a..dca2c78db5d0 100644 +--- a/arch/x86/kvm/vmx/nested.c ++++ b/arch/x86/kvm/vmx/nested.c +@@ -2074,7 +2074,7 @@ static void prepare_vmcs02_early(struct vcpu_vmx *vmx, struct vmcs12 *vmcs12) + */ + exec_control = vmx_exec_control(vmx); /* L0's desires */ + exec_control &= ~CPU_BASED_INTR_WINDOW_EXITING; +- exec_control &= ~CPU_BASED_VIRTUAL_NMI_PENDING; ++ exec_control &= ~CPU_BASED_NMI_WINDOW_EXITING; + exec_control &= ~CPU_BASED_TPR_SHADOW; + exec_control |= vmcs12->cpu_based_vm_exec_control; + +@@ -2459,7 +2459,7 @@ static int nested_vmx_check_nmi_controls(struct vmcs12 *vmcs12) + return -EINVAL; + + if (CC(!nested_cpu_has_virtual_nmis(vmcs12) && +- nested_cpu_has(vmcs12, CPU_BASED_VIRTUAL_NMI_PENDING))) ++ nested_cpu_has(vmcs12, CPU_BASED_NMI_WINDOW_EXITING))) + return -EINVAL; + + return 0; +@@ -3039,7 +3039,7 @@ enum nvmx_vmentry_status nested_vmx_enter_non_root_mode(struct kvm_vcpu *vcpu, + u32 exit_qual; + + evaluate_pending_interrupts = exec_controls_get(vmx) & +- (CPU_BASED_INTR_WINDOW_EXITING | CPU_BASED_VIRTUAL_NMI_PENDING); ++ (CPU_BASED_INTR_WINDOW_EXITING | CPU_BASED_NMI_WINDOW_EXITING); + if (likely(!evaluate_pending_interrupts) && kvm_vcpu_apicv_active(vcpu)) + evaluate_pending_interrupts |= vmx_has_apicv_interrupt(vcpu); + +@@ -3267,7 +3267,7 @@ static int nested_vmx_run(struct kvm_vcpu *vcpu, bool launch) + */ + if ((vmcs12->guest_activity_state == GUEST_ACTIVITY_HLT) && + !(vmcs12->vm_entry_intr_info_field & INTR_INFO_VALID_MASK) && +- !(vmcs12->cpu_based_vm_exec_control & CPU_BASED_VIRTUAL_NMI_PENDING) && ++ !(vmcs12->cpu_based_vm_exec_control & CPU_BASED_NMI_WINDOW_EXITING) && + !((vmcs12->cpu_based_vm_exec_control & CPU_BASED_INTR_WINDOW_EXITING) && + (vmcs12->guest_rflags & X86_EFLAGS_IF))) { + vmx->nested.nested_run_pending = 0; +@@ -5379,7 +5379,7 @@ bool nested_vmx_exit_reflected(struct kvm_vcpu *vcpu, u32 exit_reason) + case EXIT_REASON_INTERRUPT_WINDOW: + return nested_cpu_has(vmcs12, CPU_BASED_INTR_WINDOW_EXITING); + case EXIT_REASON_NMI_WINDOW: +- return nested_cpu_has(vmcs12, CPU_BASED_VIRTUAL_NMI_PENDING); ++ return nested_cpu_has(vmcs12, CPU_BASED_NMI_WINDOW_EXITING); + case EXIT_REASON_TASK_SWITCH: + return true; + case EXIT_REASON_CPUID: +@@ -5870,7 +5870,7 @@ void nested_vmx_setup_ctls_msrs(struct nested_vmx_msrs *msrs, u32 ept_caps) + CPU_BASED_ALWAYSON_WITHOUT_TRUE_MSR; + msrs->procbased_ctls_high &= + CPU_BASED_INTR_WINDOW_EXITING | +- CPU_BASED_VIRTUAL_NMI_PENDING | CPU_BASED_USE_TSC_OFFSETING | ++ CPU_BASED_NMI_WINDOW_EXITING | CPU_BASED_USE_TSC_OFFSETING | + CPU_BASED_HLT_EXITING | CPU_BASED_INVLPG_EXITING | + CPU_BASED_MWAIT_EXITING | CPU_BASED_CR3_LOAD_EXITING | + CPU_BASED_CR3_STORE_EXITING | +diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c +index 51aa5851011c..470a8f9a0046 100644 +--- a/arch/x86/kvm/vmx/vmx.c ++++ b/arch/x86/kvm/vmx/vmx.c +@@ -4469,7 +4469,7 @@ static void enable_nmi_window(struct kvm_vcpu *vcpu) + return; + } + +- exec_controls_setbit(to_vmx(vcpu), CPU_BASED_VIRTUAL_NMI_PENDING); ++ exec_controls_setbit(to_vmx(vcpu), CPU_BASED_NMI_WINDOW_EXITING); + } + + static void vmx_inject_irq(struct kvm_vcpu *vcpu) +@@ -5295,7 +5295,7 @@ static int handle_ept_misconfig(struct kvm_vcpu *vcpu) + static int handle_nmi_window(struct kvm_vcpu *vcpu) + { + WARN_ON_ONCE(!enable_vnmi); +- exec_controls_clearbit(to_vmx(vcpu), CPU_BASED_VIRTUAL_NMI_PENDING); ++ exec_controls_clearbit(to_vmx(vcpu), CPU_BASED_NMI_WINDOW_EXITING); + ++vcpu->stat.nmi_window_exits; + kvm_make_request(KVM_REQ_EVENT, vcpu); + +diff --git a/tools/testing/selftests/kvm/include/x86_64/vmx.h b/tools/testing/selftests/kvm/include/x86_64/vmx.h +index c6e442d7a241..7eb38451c359 100644 +--- a/tools/testing/selftests/kvm/include/x86_64/vmx.h ++++ b/tools/testing/selftests/kvm/include/x86_64/vmx.h +@@ -30,7 +30,7 @@ + #define CPU_BASED_CR8_LOAD_EXITING 0x00080000 + #define CPU_BASED_CR8_STORE_EXITING 0x00100000 + #define CPU_BASED_TPR_SHADOW 0x00200000 +-#define CPU_BASED_VIRTUAL_NMI_PENDING 0x00400000 ++#define CPU_BASED_NMI_WINDOW_EXITING 0x00400000 + #define CPU_BASED_MOV_DR_EXITING 0x00800000 + #define CPU_BASED_UNCOND_IO_EXITING 0x01000000 + #define CPU_BASED_USE_IO_BITMAPS 0x02000000 +-- +2.35.1 + diff --git a/queue-5.4/kvm-x86-optimize-more-exit-handlers-in-vmx.c.patch b/queue-5.4/kvm-x86-optimize-more-exit-handlers-in-vmx.c.patch new file mode 100644 index 00000000000..73d521da317 --- /dev/null +++ b/queue-5.4/kvm-x86-optimize-more-exit-handlers-in-vmx.c.patch @@ -0,0 +1,86 @@ +From ee16a936986468f7909687e3530a8912664707e2 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 4 Nov 2019 17:59:58 -0500 +Subject: KVM: x86: optimize more exit handlers in vmx.c + +From: Andrea Arcangeli + +[ Upstream commit f399e60c45f6b6e6ad6dfcedff1dd6386e086b0b ] + +Eliminate wasteful call/ret non RETPOLINE case and unnecessary fentry +dynamic tracing hooking points. + +Signed-off-by: Andrea Arcangeli +Signed-off-by: Paolo Bonzini +Stable-dep-of: 31de69f4eea7 ("KVM: nVMX: Properly expose ENABLE_USR_WAIT_PAUSE control to L1") +Signed-off-by: Sasha Levin +--- + arch/x86/kvm/vmx/vmx.c | 30 +++++------------------------- + 1 file changed, 5 insertions(+), 25 deletions(-) + +diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c +index 52f024eeac3d..0fae9b448ab9 100644 +--- a/arch/x86/kvm/vmx/vmx.c ++++ b/arch/x86/kvm/vmx/vmx.c +@@ -4802,7 +4802,7 @@ static int handle_exception_nmi(struct kvm_vcpu *vcpu) + return 0; + } + +-static int handle_external_interrupt(struct kvm_vcpu *vcpu) ++static __always_inline int handle_external_interrupt(struct kvm_vcpu *vcpu) + { + ++vcpu->stat.irq_exits; + return 1; +@@ -5074,21 +5074,6 @@ static void vmx_set_dr7(struct kvm_vcpu *vcpu, unsigned long val) + vmcs_writel(GUEST_DR7, val); + } + +-static int handle_cpuid(struct kvm_vcpu *vcpu) +-{ +- return kvm_emulate_cpuid(vcpu); +-} +- +-static int handle_rdmsr(struct kvm_vcpu *vcpu) +-{ +- return kvm_emulate_rdmsr(vcpu); +-} +- +-static int handle_wrmsr(struct kvm_vcpu *vcpu) +-{ +- return kvm_emulate_wrmsr(vcpu); +-} +- + static int handle_tpr_below_threshold(struct kvm_vcpu *vcpu) + { + kvm_apic_update_ppr(vcpu); +@@ -5105,11 +5090,6 @@ static int handle_interrupt_window(struct kvm_vcpu *vcpu) + return 1; + } + +-static int handle_halt(struct kvm_vcpu *vcpu) +-{ +- return kvm_emulate_halt(vcpu); +-} +- + static int handle_vmcall(struct kvm_vcpu *vcpu) + { + return kvm_emulate_hypercall(vcpu); +@@ -5657,11 +5637,11 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = { + [EXIT_REASON_IO_INSTRUCTION] = handle_io, + [EXIT_REASON_CR_ACCESS] = handle_cr, + [EXIT_REASON_DR_ACCESS] = handle_dr, +- [EXIT_REASON_CPUID] = handle_cpuid, +- [EXIT_REASON_MSR_READ] = handle_rdmsr, +- [EXIT_REASON_MSR_WRITE] = handle_wrmsr, ++ [EXIT_REASON_CPUID] = kvm_emulate_cpuid, ++ [EXIT_REASON_MSR_READ] = kvm_emulate_rdmsr, ++ [EXIT_REASON_MSR_WRITE] = kvm_emulate_wrmsr, + [EXIT_REASON_PENDING_INTERRUPT] = handle_interrupt_window, +- [EXIT_REASON_HLT] = handle_halt, ++ [EXIT_REASON_HLT] = kvm_emulate_halt, + [EXIT_REASON_INVD] = handle_invd, + [EXIT_REASON_INVLPG] = handle_invlpg, + [EXIT_REASON_RDPMC] = handle_rdpmc, +-- +2.35.1 + diff --git a/queue-5.4/mbcache-add-functions-to-delete-entry-if-unused.patch b/queue-5.4/mbcache-add-functions-to-delete-entry-if-unused.patch new file mode 100644 index 00000000000..2fa7b3fba1b --- /dev/null +++ b/queue-5.4/mbcache-add-functions-to-delete-entry-if-unused.patch @@ -0,0 +1,156 @@ +From 432f1ccf6545771425ca6cca6e0c1914e0b398c6 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 12 Jul 2022 12:54:21 +0200 +Subject: mbcache: add functions to delete entry if unused + +From: Jan Kara + +[ Upstream commit 3dc96bba65f53daa217f0a8f43edad145286a8f5 ] + +Add function mb_cache_entry_delete_or_get() to delete mbcache entry if +it is unused and also add a function to wait for entry to become unused +- mb_cache_entry_wait_unused(). We do not share code between the two +deleting function as one of them will go away soon. + +CC: stable@vger.kernel.org +Fixes: 82939d7999df ("ext4: convert to mbcache2") +Signed-off-by: Jan Kara +Link: https://lore.kernel.org/r/20220712105436.32204-2-jack@suse.cz +Signed-off-by: Theodore Ts'o +Stable-dep-of: a44e84a9b776 ("ext4: fix deadlock due to mbcache entry corruption") +Signed-off-by: Sasha Levin +--- + fs/mbcache.c | 66 +++++++++++++++++++++++++++++++++++++++-- + include/linux/mbcache.h | 10 ++++++- + 2 files changed, 73 insertions(+), 3 deletions(-) + +diff --git a/fs/mbcache.c b/fs/mbcache.c +index cfc28129fb6f..2010bc80a3f2 100644 +--- a/fs/mbcache.c ++++ b/fs/mbcache.c +@@ -11,7 +11,7 @@ + /* + * Mbcache is a simple key-value store. Keys need not be unique, however + * key-value pairs are expected to be unique (we use this fact in +- * mb_cache_entry_delete()). ++ * mb_cache_entry_delete_or_get()). + * + * Ext2 and ext4 use this cache for deduplication of extended attribute blocks. + * Ext4 also uses it for deduplication of xattr values stored in inodes. +@@ -125,6 +125,19 @@ void __mb_cache_entry_free(struct mb_cache_entry *entry) + } + EXPORT_SYMBOL(__mb_cache_entry_free); + ++/* ++ * mb_cache_entry_wait_unused - wait to be the last user of the entry ++ * ++ * @entry - entry to work on ++ * ++ * Wait to be the last user of the entry. ++ */ ++void mb_cache_entry_wait_unused(struct mb_cache_entry *entry) ++{ ++ wait_var_event(&entry->e_refcnt, atomic_read(&entry->e_refcnt) <= 3); ++} ++EXPORT_SYMBOL(mb_cache_entry_wait_unused); ++ + static struct mb_cache_entry *__entry_find(struct mb_cache *cache, + struct mb_cache_entry *entry, + u32 key) +@@ -217,7 +230,7 @@ struct mb_cache_entry *mb_cache_entry_get(struct mb_cache *cache, u32 key, + } + EXPORT_SYMBOL(mb_cache_entry_get); + +-/* mb_cache_entry_delete - remove a cache entry ++/* mb_cache_entry_delete - try to remove a cache entry + * @cache - cache we work with + * @key - key + * @value - value +@@ -254,6 +267,55 @@ void mb_cache_entry_delete(struct mb_cache *cache, u32 key, u64 value) + } + EXPORT_SYMBOL(mb_cache_entry_delete); + ++/* mb_cache_entry_delete_or_get - remove a cache entry if it has no users ++ * @cache - cache we work with ++ * @key - key ++ * @value - value ++ * ++ * Remove entry from cache @cache with key @key and value @value. The removal ++ * happens only if the entry is unused. The function returns NULL in case the ++ * entry was successfully removed or there's no entry in cache. Otherwise the ++ * function grabs reference of the entry that we failed to delete because it ++ * still has users and return it. ++ */ ++struct mb_cache_entry *mb_cache_entry_delete_or_get(struct mb_cache *cache, ++ u32 key, u64 value) ++{ ++ struct hlist_bl_node *node; ++ struct hlist_bl_head *head; ++ struct mb_cache_entry *entry; ++ ++ head = mb_cache_entry_head(cache, key); ++ hlist_bl_lock(head); ++ hlist_bl_for_each_entry(entry, node, head, e_hash_list) { ++ if (entry->e_key == key && entry->e_value == value) { ++ if (atomic_read(&entry->e_refcnt) > 2) { ++ atomic_inc(&entry->e_refcnt); ++ hlist_bl_unlock(head); ++ return entry; ++ } ++ /* We keep hash list reference to keep entry alive */ ++ hlist_bl_del_init(&entry->e_hash_list); ++ hlist_bl_unlock(head); ++ spin_lock(&cache->c_list_lock); ++ if (!list_empty(&entry->e_list)) { ++ list_del_init(&entry->e_list); ++ if (!WARN_ONCE(cache->c_entry_count == 0, ++ "mbcache: attempt to decrement c_entry_count past zero")) ++ cache->c_entry_count--; ++ atomic_dec(&entry->e_refcnt); ++ } ++ spin_unlock(&cache->c_list_lock); ++ mb_cache_entry_put(cache, entry); ++ return NULL; ++ } ++ } ++ hlist_bl_unlock(head); ++ ++ return NULL; ++} ++EXPORT_SYMBOL(mb_cache_entry_delete_or_get); ++ + /* mb_cache_entry_touch - cache entry got used + * @cache - cache the entry belongs to + * @entry - entry that got used +diff --git a/include/linux/mbcache.h b/include/linux/mbcache.h +index 20f1e3ff6013..8eca7f25c432 100644 +--- a/include/linux/mbcache.h ++++ b/include/linux/mbcache.h +@@ -30,15 +30,23 @@ void mb_cache_destroy(struct mb_cache *cache); + int mb_cache_entry_create(struct mb_cache *cache, gfp_t mask, u32 key, + u64 value, bool reusable); + void __mb_cache_entry_free(struct mb_cache_entry *entry); ++void mb_cache_entry_wait_unused(struct mb_cache_entry *entry); + static inline int mb_cache_entry_put(struct mb_cache *cache, + struct mb_cache_entry *entry) + { +- if (!atomic_dec_and_test(&entry->e_refcnt)) ++ unsigned int cnt = atomic_dec_return(&entry->e_refcnt); ++ ++ if (cnt > 0) { ++ if (cnt <= 3) ++ wake_up_var(&entry->e_refcnt); + return 0; ++ } + __mb_cache_entry_free(entry); + return 1; + } + ++struct mb_cache_entry *mb_cache_entry_delete_or_get(struct mb_cache *cache, ++ u32 key, u64 value); + void mb_cache_entry_delete(struct mb_cache *cache, u32 key, u64 value); + struct mb_cache_entry *mb_cache_entry_get(struct mb_cache *cache, u32 key, + u64 value); +-- +2.35.1 + diff --git a/queue-5.4/mbcache-automatically-delete-entries-from-cache-on-f.patch b/queue-5.4/mbcache-automatically-delete-entries-from-cache-on-f.patch new file mode 100644 index 00000000000..2a1d666a43a --- /dev/null +++ b/queue-5.4/mbcache-automatically-delete-entries-from-cache-on-f.patch @@ -0,0 +1,274 @@ +From c7d9d3671c90cfdcc47f373a2da48a51535960bb Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 12 Jul 2022 12:54:29 +0200 +Subject: mbcache: automatically delete entries from cache on freeing + +From: Jan Kara + +[ Upstream commit 307af6c879377c1c63e71cbdd978201f9c7ee8df ] + +Use the fact that entries with elevated refcount are not removed from +the hash and just move removal of the entry from the hash to the entry +freeing time. When doing this we also change the generic code to hold +one reference to the cache entry, not two of them, which makes code +somewhat more obvious. + +Signed-off-by: Jan Kara +Link: https://lore.kernel.org/r/20220712105436.32204-10-jack@suse.cz +Signed-off-by: Theodore Ts'o +Stable-dep-of: a44e84a9b776 ("ext4: fix deadlock due to mbcache entry corruption") +Signed-off-by: Sasha Levin +--- + fs/mbcache.c | 108 +++++++++++++++------------------------- + include/linux/mbcache.h | 24 ++++++--- + 2 files changed, 55 insertions(+), 77 deletions(-) + +diff --git a/fs/mbcache.c b/fs/mbcache.c +index 2010bc80a3f2..950f1829a7fd 100644 +--- a/fs/mbcache.c ++++ b/fs/mbcache.c +@@ -90,7 +90,7 @@ int mb_cache_entry_create(struct mb_cache *cache, gfp_t mask, u32 key, + return -ENOMEM; + + INIT_LIST_HEAD(&entry->e_list); +- /* One ref for hash, one ref returned */ ++ /* Initial hash reference */ + atomic_set(&entry->e_refcnt, 1); + entry->e_key = key; + entry->e_value = value; +@@ -106,21 +106,28 @@ int mb_cache_entry_create(struct mb_cache *cache, gfp_t mask, u32 key, + } + } + hlist_bl_add_head(&entry->e_hash_list, head); +- hlist_bl_unlock(head); +- ++ /* ++ * Add entry to LRU list before it can be found by ++ * mb_cache_entry_delete() to avoid races ++ */ + spin_lock(&cache->c_list_lock); + list_add_tail(&entry->e_list, &cache->c_list); +- /* Grab ref for LRU list */ +- atomic_inc(&entry->e_refcnt); + cache->c_entry_count++; + spin_unlock(&cache->c_list_lock); ++ hlist_bl_unlock(head); + + return 0; + } + EXPORT_SYMBOL(mb_cache_entry_create); + +-void __mb_cache_entry_free(struct mb_cache_entry *entry) ++void __mb_cache_entry_free(struct mb_cache *cache, struct mb_cache_entry *entry) + { ++ struct hlist_bl_head *head; ++ ++ head = mb_cache_entry_head(cache, entry->e_key); ++ hlist_bl_lock(head); ++ hlist_bl_del(&entry->e_hash_list); ++ hlist_bl_unlock(head); + kmem_cache_free(mb_entry_cache, entry); + } + EXPORT_SYMBOL(__mb_cache_entry_free); +@@ -134,7 +141,7 @@ EXPORT_SYMBOL(__mb_cache_entry_free); + */ + void mb_cache_entry_wait_unused(struct mb_cache_entry *entry) + { +- wait_var_event(&entry->e_refcnt, atomic_read(&entry->e_refcnt) <= 3); ++ wait_var_event(&entry->e_refcnt, atomic_read(&entry->e_refcnt) <= 2); + } + EXPORT_SYMBOL(mb_cache_entry_wait_unused); + +@@ -155,10 +162,9 @@ static struct mb_cache_entry *__entry_find(struct mb_cache *cache, + while (node) { + entry = hlist_bl_entry(node, struct mb_cache_entry, + e_hash_list); +- if (entry->e_key == key && entry->e_reusable) { +- atomic_inc(&entry->e_refcnt); ++ if (entry->e_key == key && entry->e_reusable && ++ atomic_inc_not_zero(&entry->e_refcnt)) + goto out; +- } + node = node->next; + } + entry = NULL; +@@ -218,10 +224,9 @@ struct mb_cache_entry *mb_cache_entry_get(struct mb_cache *cache, u32 key, + head = mb_cache_entry_head(cache, key); + hlist_bl_lock(head); + hlist_bl_for_each_entry(entry, node, head, e_hash_list) { +- if (entry->e_key == key && entry->e_value == value) { +- atomic_inc(&entry->e_refcnt); ++ if (entry->e_key == key && entry->e_value == value && ++ atomic_inc_not_zero(&entry->e_refcnt)) + goto out; +- } + } + entry = NULL; + out: +@@ -281,37 +286,25 @@ EXPORT_SYMBOL(mb_cache_entry_delete); + struct mb_cache_entry *mb_cache_entry_delete_or_get(struct mb_cache *cache, + u32 key, u64 value) + { +- struct hlist_bl_node *node; +- struct hlist_bl_head *head; + struct mb_cache_entry *entry; + +- head = mb_cache_entry_head(cache, key); +- hlist_bl_lock(head); +- hlist_bl_for_each_entry(entry, node, head, e_hash_list) { +- if (entry->e_key == key && entry->e_value == value) { +- if (atomic_read(&entry->e_refcnt) > 2) { +- atomic_inc(&entry->e_refcnt); +- hlist_bl_unlock(head); +- return entry; +- } +- /* We keep hash list reference to keep entry alive */ +- hlist_bl_del_init(&entry->e_hash_list); +- hlist_bl_unlock(head); +- spin_lock(&cache->c_list_lock); +- if (!list_empty(&entry->e_list)) { +- list_del_init(&entry->e_list); +- if (!WARN_ONCE(cache->c_entry_count == 0, +- "mbcache: attempt to decrement c_entry_count past zero")) +- cache->c_entry_count--; +- atomic_dec(&entry->e_refcnt); +- } +- spin_unlock(&cache->c_list_lock); +- mb_cache_entry_put(cache, entry); +- return NULL; +- } +- } +- hlist_bl_unlock(head); ++ entry = mb_cache_entry_get(cache, key, value); ++ if (!entry) ++ return NULL; + ++ /* ++ * Drop the ref we got from mb_cache_entry_get() and the initial hash ++ * ref if we are the last user ++ */ ++ if (atomic_cmpxchg(&entry->e_refcnt, 2, 0) != 2) ++ return entry; ++ ++ spin_lock(&cache->c_list_lock); ++ if (!list_empty(&entry->e_list)) ++ list_del_init(&entry->e_list); ++ cache->c_entry_count--; ++ spin_unlock(&cache->c_list_lock); ++ __mb_cache_entry_free(cache, entry); + return NULL; + } + EXPORT_SYMBOL(mb_cache_entry_delete_or_get); +@@ -343,42 +336,24 @@ static unsigned long mb_cache_shrink(struct mb_cache *cache, + unsigned long nr_to_scan) + { + struct mb_cache_entry *entry; +- struct hlist_bl_head *head; + unsigned long shrunk = 0; + + spin_lock(&cache->c_list_lock); + while (nr_to_scan-- && !list_empty(&cache->c_list)) { + entry = list_first_entry(&cache->c_list, + struct mb_cache_entry, e_list); +- if (entry->e_referenced || atomic_read(&entry->e_refcnt) > 2) { ++ /* Drop initial hash reference if there is no user */ ++ if (entry->e_referenced || ++ atomic_cmpxchg(&entry->e_refcnt, 1, 0) != 1) { + entry->e_referenced = 0; + list_move_tail(&entry->e_list, &cache->c_list); + continue; + } + list_del_init(&entry->e_list); + cache->c_entry_count--; +- /* +- * We keep LRU list reference so that entry doesn't go away +- * from under us. +- */ + spin_unlock(&cache->c_list_lock); +- head = mb_cache_entry_head(cache, entry->e_key); +- hlist_bl_lock(head); +- /* Now a reliable check if the entry didn't get used... */ +- if (atomic_read(&entry->e_refcnt) > 2) { +- hlist_bl_unlock(head); +- spin_lock(&cache->c_list_lock); +- list_add_tail(&entry->e_list, &cache->c_list); +- cache->c_entry_count++; +- continue; +- } +- if (!hlist_bl_unhashed(&entry->e_hash_list)) { +- hlist_bl_del_init(&entry->e_hash_list); +- atomic_dec(&entry->e_refcnt); +- } +- hlist_bl_unlock(head); +- if (mb_cache_entry_put(cache, entry)) +- shrunk++; ++ __mb_cache_entry_free(cache, entry); ++ shrunk++; + cond_resched(); + spin_lock(&cache->c_list_lock); + } +@@ -470,11 +445,6 @@ void mb_cache_destroy(struct mb_cache *cache) + * point. + */ + list_for_each_entry_safe(entry, next, &cache->c_list, e_list) { +- if (!hlist_bl_unhashed(&entry->e_hash_list)) { +- hlist_bl_del_init(&entry->e_hash_list); +- atomic_dec(&entry->e_refcnt); +- } else +- WARN_ON(1); + list_del(&entry->e_list); + WARN_ON(atomic_read(&entry->e_refcnt) != 1); + mb_cache_entry_put(cache, entry); +diff --git a/include/linux/mbcache.h b/include/linux/mbcache.h +index 8eca7f25c432..e9d5ece87794 100644 +--- a/include/linux/mbcache.h ++++ b/include/linux/mbcache.h +@@ -13,8 +13,16 @@ struct mb_cache; + struct mb_cache_entry { + /* List of entries in cache - protected by cache->c_list_lock */ + struct list_head e_list; +- /* Hash table list - protected by hash chain bitlock */ ++ /* ++ * Hash table list - protected by hash chain bitlock. The entry is ++ * guaranteed to be hashed while e_refcnt > 0. ++ */ + struct hlist_bl_node e_hash_list; ++ /* ++ * Entry refcount. Once it reaches zero, entry is unhashed and freed. ++ * While refcount > 0, the entry is guaranteed to stay in the hash and ++ * e.g. mb_cache_entry_try_delete() will fail. ++ */ + atomic_t e_refcnt; + /* Key in hash - stable during lifetime of the entry */ + u32 e_key; +@@ -29,20 +37,20 @@ void mb_cache_destroy(struct mb_cache *cache); + + int mb_cache_entry_create(struct mb_cache *cache, gfp_t mask, u32 key, + u64 value, bool reusable); +-void __mb_cache_entry_free(struct mb_cache_entry *entry); ++void __mb_cache_entry_free(struct mb_cache *cache, ++ struct mb_cache_entry *entry); + void mb_cache_entry_wait_unused(struct mb_cache_entry *entry); +-static inline int mb_cache_entry_put(struct mb_cache *cache, +- struct mb_cache_entry *entry) ++static inline void mb_cache_entry_put(struct mb_cache *cache, ++ struct mb_cache_entry *entry) + { + unsigned int cnt = atomic_dec_return(&entry->e_refcnt); + + if (cnt > 0) { +- if (cnt <= 3) ++ if (cnt <= 2) + wake_up_var(&entry->e_refcnt); +- return 0; ++ return; + } +- __mb_cache_entry_free(entry); +- return 1; ++ __mb_cache_entry_free(cache, entry); + } + + struct mb_cache_entry *mb_cache_entry_delete_or_get(struct mb_cache *cache, +-- +2.35.1 + diff --git a/queue-5.4/mbcache-don-t-reclaim-used-entries.patch b/queue-5.4/mbcache-don-t-reclaim-used-entries.patch new file mode 100644 index 00000000000..e70b6696b8d --- /dev/null +++ b/queue-5.4/mbcache-don-t-reclaim-used-entries.patch @@ -0,0 +1,56 @@ +From 3fdd29921b1dd5c928461ad78613944700e56e38 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 12 Jul 2022 12:54:20 +0200 +Subject: mbcache: don't reclaim used entries + +From: Jan Kara + +[ Upstream commit 58318914186c157477b978b1739dfe2f1b9dc0fe ] + +Do not reclaim entries that are currently used by somebody from a +shrinker. Firstly, these entries are likely useful. Secondly, we will +need to keep such entries to protect pending increment of xattr block +refcount. + +CC: stable@vger.kernel.org +Fixes: 82939d7999df ("ext4: convert to mbcache2") +Signed-off-by: Jan Kara +Link: https://lore.kernel.org/r/20220712105436.32204-1-jack@suse.cz +Signed-off-by: Theodore Ts'o +Stable-dep-of: a44e84a9b776 ("ext4: fix deadlock due to mbcache entry corruption") +Signed-off-by: Sasha Levin +--- + fs/mbcache.c | 10 +++++++++- + 1 file changed, 9 insertions(+), 1 deletion(-) + +diff --git a/fs/mbcache.c b/fs/mbcache.c +index 97c54d3a2227..cfc28129fb6f 100644 +--- a/fs/mbcache.c ++++ b/fs/mbcache.c +@@ -288,7 +288,7 @@ static unsigned long mb_cache_shrink(struct mb_cache *cache, + while (nr_to_scan-- && !list_empty(&cache->c_list)) { + entry = list_first_entry(&cache->c_list, + struct mb_cache_entry, e_list); +- if (entry->e_referenced) { ++ if (entry->e_referenced || atomic_read(&entry->e_refcnt) > 2) { + entry->e_referenced = 0; + list_move_tail(&entry->e_list, &cache->c_list); + continue; +@@ -302,6 +302,14 @@ static unsigned long mb_cache_shrink(struct mb_cache *cache, + spin_unlock(&cache->c_list_lock); + head = mb_cache_entry_head(cache, entry->e_key); + hlist_bl_lock(head); ++ /* Now a reliable check if the entry didn't get used... */ ++ if (atomic_read(&entry->e_refcnt) > 2) { ++ hlist_bl_unlock(head); ++ spin_lock(&cache->c_list_lock); ++ list_add_tail(&entry->e_list, &cache->c_list); ++ cache->c_entry_count++; ++ continue; ++ } + if (!hlist_bl_unhashed(&entry->e_hash_list)) { + hlist_bl_del_init(&entry->e_hash_list); + atomic_dec(&entry->e_refcnt); +-- +2.35.1 + diff --git a/queue-5.4/mm-highmem-lift-memcpy_-to-from-_page-to-core.patch b/queue-5.4/mm-highmem-lift-memcpy_-to-from-_page-to-core.patch new file mode 100644 index 00000000000..271ba885d59 --- /dev/null +++ b/queue-5.4/mm-highmem-lift-memcpy_-to-from-_page-to-core.patch @@ -0,0 +1,123 @@ +From 6a652141db3eebe1e2ee16009210cb3d4e76d78c Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 9 Feb 2021 22:22:14 -0800 +Subject: mm/highmem: Lift memcpy_[to|from]_page to core + +From: Ira Weiny + +[ Upstream commit bb90d4bc7b6a536b2e4db45f4763e467c2008251 ] + +Working through a conversion to a call kmap_local_page() instead of +kmap() revealed many places where the pattern kmap/memcpy/kunmap +occurred. + +Eric Biggers, Matthew Wilcox, Christoph Hellwig, Dan Williams, and Al +Viro all suggested putting this code into helper functions. Al Viro +further pointed out that these functions already existed in the iov_iter +code.[1] + +Various locations for the lifted functions were considered. + +Headers like mm.h or string.h seem ok but don't really portray the +functionality well. pagemap.h made some sense but is for page cache +functionality.[2] + +Another alternative would be to create a new header for the promoted +memcpy functions, but it masks the fact that these are designed to copy +to/from pages using the kernel direct mappings and complicates matters +with a new header. + +Placing these functions in 'highmem.h' is suboptimal especially with the +changes being proposed in the functionality of kmap. From a caller +perspective including/using 'highmem.h' implies that the functions +defined in that header are only required when highmem is in use which is +increasingly not the case with modern processors. However, highmem.h is +where all the current functions like this reside (zero_user(), +clear_highpage(), clear_user_highpage(), copy_user_highpage(), and +copy_highpage()). So it makes the most sense even though it is +distasteful for some.[3] + +Lift memcpy_to_page() and memcpy_from_page() to pagemap.h. + +[1] https://lore.kernel.org/lkml/20201013200149.GI3576660@ZenIV.linux.org.uk/ + https://lore.kernel.org/lkml/20201013112544.GA5249@infradead.org/ + +[2] https://lore.kernel.org/lkml/20201208122316.GH7338@casper.infradead.org/ + +[3] https://lore.kernel.org/lkml/20201013200149.GI3576660@ZenIV.linux.org.uk/#t + https://lore.kernel.org/lkml/20201208163814.GN1563847@iweiny-DESK2.sc.intel.com/ + +Cc: Boris Pismenny +Cc: Or Gerlitz +Cc: Dave Hansen +Suggested-by: Matthew Wilcox +Suggested-by: Christoph Hellwig +Suggested-by: Dan Williams +Suggested-by: Al Viro +Suggested-by: Eric Biggers +Reviewed-by: Chaitanya Kulkarni +Reviewed-by: Christoph Hellwig +Signed-off-by: Ira Weiny +Signed-off-by: David Sterba +Stable-dep-of: 956510c0c743 ("fs: ext4: initialize fsdata in pagecache_write()") +Signed-off-by: Sasha Levin +--- + include/linux/highmem.h | 18 ++++++++++++++++++ + lib/iov_iter.c | 14 -------------- + 2 files changed, 18 insertions(+), 14 deletions(-) + +diff --git a/include/linux/highmem.h b/include/linux/highmem.h +index ea5cdbd8c2c3..900f224bb640 100644 +--- a/include/linux/highmem.h ++++ b/include/linux/highmem.h +@@ -276,4 +276,22 @@ static inline void copy_highpage(struct page *to, struct page *from) + + #endif + ++static inline void memcpy_from_page(char *to, struct page *page, ++ size_t offset, size_t len) ++{ ++ char *from = kmap_atomic(page); ++ ++ memcpy(to, from + offset, len); ++ kunmap_atomic(from); ++} ++ ++static inline void memcpy_to_page(struct page *page, size_t offset, ++ const char *from, size_t len) ++{ ++ char *to = kmap_atomic(page); ++ ++ memcpy(to + offset, from, len); ++ kunmap_atomic(to); ++} ++ + #endif /* _LINUX_HIGHMEM_H */ +diff --git a/lib/iov_iter.c b/lib/iov_iter.c +index 9d3bda3d49fe..5c6a0b8a2adb 100644 +--- a/lib/iov_iter.c ++++ b/lib/iov_iter.c +@@ -455,20 +455,6 @@ void iov_iter_init(struct iov_iter *i, unsigned int direction, + } + EXPORT_SYMBOL(iov_iter_init); + +-static void memcpy_from_page(char *to, struct page *page, size_t offset, size_t len) +-{ +- char *from = kmap_atomic(page); +- memcpy(to, from + offset, len); +- kunmap_atomic(from); +-} +- +-static void memcpy_to_page(struct page *page, size_t offset, const char *from, size_t len) +-{ +- char *to = kmap_atomic(page); +- memcpy(to + offset, from, len); +- kunmap_atomic(to); +-} +- + static void memzero_page(struct page *page, size_t offset, size_t len) + { + char *addr = kmap_atomic(page); +-- +2.35.1 + diff --git a/queue-5.4/net-amd-xgbe-add-missed-tasklet_kill.patch b/queue-5.4/net-amd-xgbe-add-missed-tasklet_kill.patch new file mode 100644 index 00000000000..37a16a9e69e --- /dev/null +++ b/queue-5.4/net-amd-xgbe-add-missed-tasklet_kill.patch @@ -0,0 +1,71 @@ +From 0c7034ff703597cb436ff4afd9c3b0037ed5837e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 28 Dec 2022 16:14:47 +0800 +Subject: net: amd-xgbe: add missed tasklet_kill + +From: Jiguang Xiao + +[ Upstream commit d530ece70f16f912e1d1bfeea694246ab78b0a4b ] + +The driver does not call tasklet_kill in several places. +Add the calls to fix it. + +Fixes: 85b85c853401 ("amd-xgbe: Re-issue interrupt if interrupt status not cleared") +Signed-off-by: Jiguang Xiao +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/amd/xgbe/xgbe-drv.c | 3 +++ + drivers/net/ethernet/amd/xgbe/xgbe-i2c.c | 4 +++- + drivers/net/ethernet/amd/xgbe/xgbe-mdio.c | 4 +++- + 3 files changed, 9 insertions(+), 2 deletions(-) + +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +index 0442d7e1cd20..7f705483c1c5 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe-drv.c ++++ b/drivers/net/ethernet/amd/xgbe/xgbe-drv.c +@@ -1139,6 +1139,9 @@ static void xgbe_free_irqs(struct xgbe_prv_data *pdata) + + devm_free_irq(pdata->dev, pdata->dev_irq, pdata); + ++ tasklet_kill(&pdata->tasklet_dev); ++ tasklet_kill(&pdata->tasklet_ecc); ++ + if (pdata->vdata->ecc_support && (pdata->dev_irq != pdata->ecc_irq)) + devm_free_irq(pdata->dev, pdata->ecc_irq, pdata); + +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c +index 4d9062d35930..530043742a07 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c ++++ b/drivers/net/ethernet/amd/xgbe/xgbe-i2c.c +@@ -447,8 +447,10 @@ static void xgbe_i2c_stop(struct xgbe_prv_data *pdata) + xgbe_i2c_disable(pdata); + xgbe_i2c_clear_all_interrupts(pdata); + +- if (pdata->dev_irq != pdata->i2c_irq) ++ if (pdata->dev_irq != pdata->i2c_irq) { + devm_free_irq(pdata->dev, pdata->i2c_irq, pdata); ++ tasklet_kill(&pdata->tasklet_i2c); ++ } + } + + static int xgbe_i2c_start(struct xgbe_prv_data *pdata) +diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +index 156a0bc8ab01..97167fc9bebe 100644 +--- a/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c ++++ b/drivers/net/ethernet/amd/xgbe/xgbe-mdio.c +@@ -1390,8 +1390,10 @@ static void xgbe_phy_stop(struct xgbe_prv_data *pdata) + /* Disable auto-negotiation */ + xgbe_an_disable_all(pdata); + +- if (pdata->dev_irq != pdata->an_irq) ++ if (pdata->dev_irq != pdata->an_irq) { + devm_free_irq(pdata->dev, pdata->an_irq, pdata); ++ tasklet_kill(&pdata->tasklet_an); ++ } + + pdata->phy_if.phy_impl.stop(pdata); + +-- +2.35.1 + diff --git a/queue-5.4/net-hns3-add-interrupts-re-initialization-while-doin.patch b/queue-5.4/net-hns3-add-interrupts-re-initialization-while-doin.patch new file mode 100644 index 00000000000..967ba3390a3 --- /dev/null +++ b/queue-5.4/net-hns3-add-interrupts-re-initialization-while-doin.patch @@ -0,0 +1,43 @@ +From d80916643dac2b2d3ebce525117cb05e3fde4a01 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 22 Dec 2022 14:43:41 +0800 +Subject: net: hns3: add interrupts re-initialization while doing VF FLR + +From: Jie Wang + +[ Upstream commit 09e6b30eeb254f1818a008cace3547159e908dfd ] + +Currently keep alive message between PF and VF may be lost and the VF is +unalive in PF. So the VF will not do reset during PF FLR reset process. +This would make the allocated interrupt resources of VF invalid and VF +would't receive or respond to PF any more. + +So this patch adds VF interrupts re-initialization during VF FLR for VF +recovery in above cases. + +Fixes: 862d969a3a4d ("net: hns3: do VF's pci re-initialization while PF doing FLR") +Signed-off-by: Jie Wang +Signed-off-by: Hao Lan +Signed-off-by: Jakub Kicinski +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +index 403c1b9cf6ab..48956c30d2ee 100644 +--- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c ++++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +@@ -2592,7 +2592,8 @@ static int hclgevf_pci_reset(struct hclgevf_dev *hdev) + struct pci_dev *pdev = hdev->pdev; + int ret = 0; + +- if (hdev->reset_type == HNAE3_VF_FULL_RESET && ++ if ((hdev->reset_type == HNAE3_VF_FULL_RESET || ++ hdev->reset_type == HNAE3_FLR_RESET) && + test_bit(HCLGEVF_STATE_IRQ_INITED, &hdev->state)) { + hclgevf_misc_irq_uninit(hdev); + hclgevf_uninit_msi(hdev); +-- +2.35.1 + diff --git a/queue-5.4/net-phy-xgmiitorgmii-fix-refcount-leak-in-xgmiitorgm.patch b/queue-5.4/net-phy-xgmiitorgmii-fix-refcount-leak-in-xgmiitorgm.patch new file mode 100644 index 00000000000..fb51b1236f9 --- /dev/null +++ b/queue-5.4/net-phy-xgmiitorgmii-fix-refcount-leak-in-xgmiitorgm.patch @@ -0,0 +1,35 @@ +From b3cec34c25dd6e26eb88b121ad678d0c1df2bd42 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 29 Dec 2022 10:29:25 +0400 +Subject: net: phy: xgmiitorgmii: Fix refcount leak in xgmiitorgmii_probe + +From: Miaoqian Lin + +[ Upstream commit d039535850ee47079d59527e96be18d8e0daa84b ] + +of_phy_find_device() return device node with refcount incremented. +Call put_device() to relese it when not needed anymore. + +Fixes: ab4e6ee578e8 ("net: phy: xgmiitorgmii: Check phy_driver ready before accessing") +Signed-off-by: Miaoqian Lin +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + drivers/net/phy/xilinx_gmii2rgmii.c | 1 + + 1 file changed, 1 insertion(+) + +diff --git a/drivers/net/phy/xilinx_gmii2rgmii.c b/drivers/net/phy/xilinx_gmii2rgmii.c +index 151c2a3f0b3a..7a78dfdfa5bd 100644 +--- a/drivers/net/phy/xilinx_gmii2rgmii.c ++++ b/drivers/net/phy/xilinx_gmii2rgmii.c +@@ -82,6 +82,7 @@ static int xgmiitorgmii_probe(struct mdio_device *mdiodev) + + if (!priv->phy_dev->drv) { + dev_info(dev, "Attached phy not ready\n"); ++ put_device(&priv->phy_dev->mdio.dev); + return -EPROBE_DEFER; + } + +-- +2.35.1 + diff --git a/queue-5.4/net-sched-atm-dont-intepret-cls-results-when-asked-t.patch b/queue-5.4/net-sched-atm-dont-intepret-cls-results-when-asked-t.patch new file mode 100644 index 00000000000..4eb20a9be5f --- /dev/null +++ b/queue-5.4/net-sched-atm-dont-intepret-cls-results-when-asked-t.patch @@ -0,0 +1,42 @@ +From 5e60452adf97e5eb139296d126fcbcf75c0eaed4 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sun, 1 Jan 2023 16:57:43 -0500 +Subject: net: sched: atm: dont intepret cls results when asked to drop + +From: Jamal Hadi Salim + +[ Upstream commit a2965c7be0522eaa18808684b7b82b248515511b ] + +If asked to drop a packet via TC_ACT_SHOT it is unsafe to assume +res.class contains a valid pointer +Fixes: b0188d4dbe5f ("[NET_SCHED]: sch_atm: Lindent") + +Signed-off-by: Jamal Hadi Salim +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + net/sched/sch_atm.c | 5 ++++- + 1 file changed, 4 insertions(+), 1 deletion(-) + +diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c +index 6385995dc700..34dd0434d99d 100644 +--- a/net/sched/sch_atm.c ++++ b/net/sched/sch_atm.c +@@ -396,10 +396,13 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch, + result = tcf_classify(skb, fl, &res, true); + if (result < 0) + continue; ++ if (result == TC_ACT_SHOT) ++ goto done; ++ + flow = (struct atm_flow_data *)res.class; + if (!flow) + flow = lookup_flow(sch, res.classid); +- goto done; ++ goto drop; + } + } + flow = NULL; +-- +2.35.1 + diff --git a/queue-5.4/net-sched-cbq-dont-intepret-cls-results-when-asked-t.patch b/queue-5.4/net-sched-cbq-dont-intepret-cls-results-when-asked-t.patch new file mode 100644 index 00000000000..813c40ca707 --- /dev/null +++ b/queue-5.4/net-sched-cbq-dont-intepret-cls-results-when-asked-t.patch @@ -0,0 +1,148 @@ +From 81d14728d7f39dace568263c6a7647b42c6658d8 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sun, 1 Jan 2023 16:57:44 -0500 +Subject: net: sched: cbq: dont intepret cls results when asked to drop + +From: Jamal Hadi Salim + +[ Upstream commit caa4b35b4317d5147b3ab0fbdc9c075c7d2e9c12 ] + +If asked to drop a packet via TC_ACT_SHOT it is unsafe to assume that +res.class contains a valid pointer + +Sample splat reported by Kyle Zeng + +[ 5.405624] 0: reclassify loop, rule prio 0, protocol 800 +[ 5.406326] ================================================================== +[ 5.407240] BUG: KASAN: slab-out-of-bounds in cbq_enqueue+0x54b/0xea0 +[ 5.407987] Read of size 1 at addr ffff88800e3122aa by task poc/299 +[ 5.408731] +[ 5.408897] CPU: 0 PID: 299 Comm: poc Not tainted 5.10.155+ #15 +[ 5.409516] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), +BIOS 1.15.0-1 04/01/2014 +[ 5.410439] Call Trace: +[ 5.410764] dump_stack+0x87/0xcd +[ 5.411153] print_address_description+0x7a/0x6b0 +[ 5.411687] ? vprintk_func+0xb9/0xc0 +[ 5.411905] ? printk+0x76/0x96 +[ 5.412110] ? cbq_enqueue+0x54b/0xea0 +[ 5.412323] kasan_report+0x17d/0x220 +[ 5.412591] ? cbq_enqueue+0x54b/0xea0 +[ 5.412803] __asan_report_load1_noabort+0x10/0x20 +[ 5.413119] cbq_enqueue+0x54b/0xea0 +[ 5.413400] ? __kasan_check_write+0x10/0x20 +[ 5.413679] __dev_queue_xmit+0x9c0/0x1db0 +[ 5.413922] dev_queue_xmit+0xc/0x10 +[ 5.414136] ip_finish_output2+0x8bc/0xcd0 +[ 5.414436] __ip_finish_output+0x472/0x7a0 +[ 5.414692] ip_finish_output+0x5c/0x190 +[ 5.414940] ip_output+0x2d8/0x3c0 +[ 5.415150] ? ip_mc_finish_output+0x320/0x320 +[ 5.415429] __ip_queue_xmit+0x753/0x1760 +[ 5.415664] ip_queue_xmit+0x47/0x60 +[ 5.415874] __tcp_transmit_skb+0x1ef9/0x34c0 +[ 5.416129] tcp_connect+0x1f5e/0x4cb0 +[ 5.416347] tcp_v4_connect+0xc8d/0x18c0 +[ 5.416577] __inet_stream_connect+0x1ae/0xb40 +[ 5.416836] ? local_bh_enable+0x11/0x20 +[ 5.417066] ? lock_sock_nested+0x175/0x1d0 +[ 5.417309] inet_stream_connect+0x5d/0x90 +[ 5.417548] ? __inet_stream_connect+0xb40/0xb40 +[ 5.417817] __sys_connect+0x260/0x2b0 +[ 5.418037] __x64_sys_connect+0x76/0x80 +[ 5.418267] do_syscall_64+0x31/0x50 +[ 5.418477] entry_SYSCALL_64_after_hwframe+0x61/0xc6 +[ 5.418770] RIP: 0033:0x473bb7 +[ 5.418952] Code: 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 +00 00 90 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2a 00 00 +00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 18 89 54 24 0c 48 89 34 +24 89 +[ 5.420046] RSP: 002b:00007fffd20eb0f8 EFLAGS: 00000246 ORIG_RAX: +000000000000002a +[ 5.420472] RAX: ffffffffffffffda RBX: 00007fffd20eb578 RCX: 0000000000473bb7 +[ 5.420872] RDX: 0000000000000010 RSI: 00007fffd20eb110 RDI: 0000000000000007 +[ 5.421271] RBP: 00007fffd20eb150 R08: 0000000000000001 R09: 0000000000000004 +[ 5.421671] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001 +[ 5.422071] R13: 00007fffd20eb568 R14: 00000000004fc740 R15: 0000000000000002 +[ 5.422471] +[ 5.422562] Allocated by task 299: +[ 5.422782] __kasan_kmalloc+0x12d/0x160 +[ 5.423007] kasan_kmalloc+0x5/0x10 +[ 5.423208] kmem_cache_alloc_trace+0x201/0x2e0 +[ 5.423492] tcf_proto_create+0x65/0x290 +[ 5.423721] tc_new_tfilter+0x137e/0x1830 +[ 5.423957] rtnetlink_rcv_msg+0x730/0x9f0 +[ 5.424197] netlink_rcv_skb+0x166/0x300 +[ 5.424428] rtnetlink_rcv+0x11/0x20 +[ 5.424639] netlink_unicast+0x673/0x860 +[ 5.424870] netlink_sendmsg+0x6af/0x9f0 +[ 5.425100] __sys_sendto+0x58d/0x5a0 +[ 5.425315] __x64_sys_sendto+0xda/0xf0 +[ 5.425539] do_syscall_64+0x31/0x50 +[ 5.425764] entry_SYSCALL_64_after_hwframe+0x61/0xc6 +[ 5.426065] +[ 5.426157] The buggy address belongs to the object at ffff88800e312200 +[ 5.426157] which belongs to the cache kmalloc-128 of size 128 +[ 5.426955] The buggy address is located 42 bytes to the right of +[ 5.426955] 128-byte region [ffff88800e312200, ffff88800e312280) +[ 5.427688] The buggy address belongs to the page: +[ 5.427992] page:000000009875fabc refcount:1 mapcount:0 +mapping:0000000000000000 index:0x0 pfn:0xe312 +[ 5.428562] flags: 0x100000000000200(slab) +[ 5.428812] raw: 0100000000000200 dead000000000100 dead000000000122 +ffff888007843680 +[ 5.429325] raw: 0000000000000000 0000000000100010 00000001ffffffff +ffff88800e312401 +[ 5.429875] page dumped because: kasan: bad access detected +[ 5.430214] page->mem_cgroup:ffff88800e312401 +[ 5.430471] +[ 5.430564] Memory state around the buggy address: +[ 5.430846] ffff88800e312180: fc fc fc fc fc fc fc fc fc fc fc fc +fc fc fc fc +[ 5.431267] ffff88800e312200: 00 00 00 00 00 00 00 00 00 00 00 00 +00 00 00 fc +[ 5.431705] >ffff88800e312280: fc fc fc fc fc fc fc fc fc fc fc fc +fc fc fc fc +[ 5.432123] ^ +[ 5.432391] ffff88800e312300: 00 00 00 00 00 00 00 00 00 00 00 00 +00 00 00 fc +[ 5.432810] ffff88800e312380: fc fc fc fc fc fc fc fc fc fc fc fc +fc fc fc fc +[ 5.433229] ================================================================== +[ 5.433648] Disabling lock debugging due to kernel taint + +Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") +Reported-by: Kyle Zeng +Signed-off-by: Jamal Hadi Salim +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + net/sched/sch_cbq.c | 5 +++-- + 1 file changed, 3 insertions(+), 2 deletions(-) + +diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c +index e5972889cd81..12893dac8461 100644 +--- a/net/sched/sch_cbq.c ++++ b/net/sched/sch_cbq.c +@@ -231,6 +231,8 @@ cbq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) + result = tcf_classify(skb, fl, &res, true); + if (!fl || result < 0) + goto fallback; ++ if (result == TC_ACT_SHOT) ++ return NULL; + + cl = (void *)res.class; + if (!cl) { +@@ -251,8 +253,7 @@ cbq_classify(struct sk_buff *skb, struct Qdisc *sch, int *qerr) + case TC_ACT_TRAP: + *qerr = NET_XMIT_SUCCESS | __NET_XMIT_STOLEN; + /* fall through */ +- case TC_ACT_SHOT: +- return NULL; ++ fallthrough; + case TC_ACT_RECLASSIFY: + return cbq_reclassify(skb, cl); + } +-- +2.35.1 + diff --git a/queue-5.4/net-sched-fix-memory-leak-in-tcindex_set_parms.patch b/queue-5.4/net-sched-fix-memory-leak-in-tcindex_set_parms.patch new file mode 100644 index 00000000000..f27b40cf546 --- /dev/null +++ b/queue-5.4/net-sched-fix-memory-leak-in-tcindex_set_parms.patch @@ -0,0 +1,150 @@ +From a37a00057e8cbf9cb16e562e9209f19202401081 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 22 Dec 2022 11:51:19 +0800 +Subject: net: sched: fix memory leak in tcindex_set_parms + +From: Hawkins Jiawei + +[ Upstream commit 399ab7fe0fa0d846881685fd4e57e9a8ef7559f7 ] + +Syzkaller reports a memory leak as follows: +==================================== +BUG: memory leak +unreferenced object 0xffff88810c287f00 (size 256): + comm "syz-executor105", pid 3600, jiffies 4294943292 (age 12.990s) + hex dump (first 32 bytes): + 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ + 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ + backtrace: + [] kmalloc_trace+0x20/0x90 mm/slab_common.c:1046 + [] kmalloc include/linux/slab.h:576 [inline] + [] kmalloc_array include/linux/slab.h:627 [inline] + [] kcalloc include/linux/slab.h:659 [inline] + [] tcf_exts_init include/net/pkt_cls.h:250 [inline] + [] tcindex_set_parms+0xa7/0xbe0 net/sched/cls_tcindex.c:342 + [] tcindex_change+0xdf/0x120 net/sched/cls_tcindex.c:553 + [] tc_new_tfilter+0x4f2/0x1100 net/sched/cls_api.c:2147 + [] rtnetlink_rcv_msg+0x4dc/0x5d0 net/core/rtnetlink.c:6082 + [] netlink_rcv_skb+0x87/0x1d0 net/netlink/af_netlink.c:2540 + [] netlink_unicast_kernel net/netlink/af_netlink.c:1319 [inline] + [] netlink_unicast+0x397/0x4c0 net/netlink/af_netlink.c:1345 + [] netlink_sendmsg+0x396/0x710 net/netlink/af_netlink.c:1921 + [] sock_sendmsg_nosec net/socket.c:714 [inline] + [] sock_sendmsg+0x56/0x80 net/socket.c:734 + [] ____sys_sendmsg+0x178/0x410 net/socket.c:2482 + [] ___sys_sendmsg+0xa8/0x110 net/socket.c:2536 + [] __sys_sendmmsg+0x105/0x330 net/socket.c:2622 + [] __do_sys_sendmmsg net/socket.c:2651 [inline] + [] __se_sys_sendmmsg net/socket.c:2648 [inline] + [] __x64_sys_sendmmsg+0x24/0x30 net/socket.c:2648 + [] do_syscall_x64 arch/x86/entry/common.c:50 [inline] + [] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 + [] entry_SYSCALL_64_after_hwframe+0x63/0xcd +==================================== + +Kernel uses tcindex_change() to change an existing +filter properties. + +Yet the problem is that, during the process of changing, +if `old_r` is retrieved from `p->perfect`, then +kernel uses tcindex_alloc_perfect_hash() to newly +allocate filter results, uses tcindex_filter_result_init() +to clear the old filter result, without destroying +its tcf_exts structure, which triggers the above memory leak. + +To be more specific, there are only two source for the `old_r`, +according to the tcindex_lookup(). `old_r` is retrieved from +`p->perfect`, or `old_r` is retrieved from `p->h`. + + * If `old_r` is retrieved from `p->perfect`, kernel uses +tcindex_alloc_perfect_hash() to newly allocate the +filter results. Then `r` is assigned with `cp->perfect + handle`, +which is newly allocated. So condition `old_r && old_r != r` is +true in this situation, and kernel uses tcindex_filter_result_init() +to clear the old filter result, without destroying +its tcf_exts structure + + * If `old_r` is retrieved from `p->h`, then `p->perfect` is NULL +according to the tcindex_lookup(). Considering that `cp->h` +is directly copied from `p->h` and `p->perfect` is NULL, +`r` is assigned with `tcindex_lookup(cp, handle)`, whose value +should be the same as `old_r`, so condition `old_r && old_r != r` +is false in this situation, kernel ignores using +tcindex_filter_result_init() to clear the old filter result. + +So only when `old_r` is retrieved from `p->perfect` does kernel use +tcindex_filter_result_init() to clear the old filter result, which +triggers the above memory leak. + +Considering that there already exists a tc_filter_wq workqueue +to destroy the old tcindex_data by tcindex_partial_destroy_work() +at the end of tcindex_set_parms(), this patch solves +this memory leak bug by removing this old filter result +clearing part and delegating it to the tc_filter_wq workqueue. + +Note that this patch doesn't introduce any other issues. If +`old_r` is retrieved from `p->perfect`, this patch just +delegates old filter result clearing part to the +tc_filter_wq workqueue; If `old_r` is retrieved from `p->h`, +kernel doesn't reach the old filter result clearing part, so +removing this part has no effect. + +[Thanks to the suggestion from Jakub Kicinski, Cong Wang, Paolo Abeni +and Dmitry Vyukov] + +Fixes: b9a24bb76bf6 ("net_sched: properly handle failure case of tcf_exts_init()") +Link: https://lore.kernel.org/all/0000000000001de5c505ebc9ec59@google.com/ +Reported-by: syzbot+232ebdbd36706c965ebf@syzkaller.appspotmail.com +Tested-by: syzbot+232ebdbd36706c965ebf@syzkaller.appspotmail.com +Cc: Cong Wang +Cc: Jakub Kicinski +Cc: Paolo Abeni +Cc: Dmitry Vyukov +Acked-by: Paolo Abeni +Signed-off-by: Hawkins Jiawei +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + net/sched/cls_tcindex.c | 12 ++---------- + 1 file changed, 2 insertions(+), 10 deletions(-) + +diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c +index 684187a1fdb9..768cf7cf65b4 100644 +--- a/net/sched/cls_tcindex.c ++++ b/net/sched/cls_tcindex.c +@@ -332,7 +332,7 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, + struct tcindex_filter_result *r, struct nlattr **tb, + struct nlattr *est, bool ovr, struct netlink_ext_ack *extack) + { +- struct tcindex_filter_result new_filter_result, *old_r = r; ++ struct tcindex_filter_result new_filter_result; + struct tcindex_data *cp = NULL, *oldp; + struct tcindex_filter *f = NULL; /* make gcc behave */ + struct tcf_result cr = {}; +@@ -401,7 +401,7 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, + err = tcindex_filter_result_init(&new_filter_result, cp, net); + if (err < 0) + goto errout_alloc; +- if (old_r) ++ if (r) + cr = r->res; + + err = -EBUSY; +@@ -478,14 +478,6 @@ tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base, + tcf_bind_filter(tp, &cr, base); + } + +- if (old_r && old_r != r) { +- err = tcindex_filter_result_init(old_r, cp, net); +- if (err < 0) { +- kfree(f); +- goto errout_alloc; +- } +- } +- + oldp = p; + r->res = cr; + tcf_exts_change(&r->exts, &e); +-- +2.35.1 + diff --git a/queue-5.4/nfc-fix-potential-resource-leaks.patch b/queue-5.4/nfc-fix-potential-resource-leaks.patch new file mode 100644 index 00000000000..b1928014793 --- /dev/null +++ b/queue-5.4/nfc-fix-potential-resource-leaks.patch @@ -0,0 +1,127 @@ +From 7556ba6b14b5120103548a4fbbe05f63e3287b06 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Fri, 23 Dec 2022 11:37:18 +0400 +Subject: nfc: Fix potential resource leaks + +From: Miaoqian Lin + +[ Upstream commit df49908f3c52d211aea5e2a14a93bbe67a2cb3af ] + +nfc_get_device() take reference for the device, add missing +nfc_put_device() to release it when not need anymore. +Also fix the style warnning by use error EOPNOTSUPP instead of +ENOTSUPP. + +Fixes: 5ce3f32b5264 ("NFC: netlink: SE API implementation") +Fixes: 29e76924cf08 ("nfc: netlink: Add capability to reply to vendor_cmd with data") +Signed-off-by: Miaoqian Lin +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + net/nfc/netlink.c | 52 ++++++++++++++++++++++++++++++++++------------- + 1 file changed, 38 insertions(+), 14 deletions(-) + +diff --git a/net/nfc/netlink.c b/net/nfc/netlink.c +index 9e94f732e717..b53d5eb86864 100644 +--- a/net/nfc/netlink.c ++++ b/net/nfc/netlink.c +@@ -1505,6 +1505,7 @@ static int nfc_genl_se_io(struct sk_buff *skb, struct genl_info *info) + u32 dev_idx, se_idx; + u8 *apdu; + size_t apdu_len; ++ int rc; + + if (!info->attrs[NFC_ATTR_DEVICE_INDEX] || + !info->attrs[NFC_ATTR_SE_INDEX] || +@@ -1518,25 +1519,37 @@ static int nfc_genl_se_io(struct sk_buff *skb, struct genl_info *info) + if (!dev) + return -ENODEV; + +- if (!dev->ops || !dev->ops->se_io) +- return -ENOTSUPP; ++ if (!dev->ops || !dev->ops->se_io) { ++ rc = -EOPNOTSUPP; ++ goto put_dev; ++ } + + apdu_len = nla_len(info->attrs[NFC_ATTR_SE_APDU]); +- if (apdu_len == 0) +- return -EINVAL; ++ if (apdu_len == 0) { ++ rc = -EINVAL; ++ goto put_dev; ++ } + + apdu = nla_data(info->attrs[NFC_ATTR_SE_APDU]); +- if (!apdu) +- return -EINVAL; ++ if (!apdu) { ++ rc = -EINVAL; ++ goto put_dev; ++ } + + ctx = kzalloc(sizeof(struct se_io_ctx), GFP_KERNEL); +- if (!ctx) +- return -ENOMEM; ++ if (!ctx) { ++ rc = -ENOMEM; ++ goto put_dev; ++ } + + ctx->dev_idx = dev_idx; + ctx->se_idx = se_idx; + +- return nfc_se_io(dev, se_idx, apdu, apdu_len, se_io_cb, ctx); ++ rc = nfc_se_io(dev, se_idx, apdu, apdu_len, se_io_cb, ctx); ++ ++put_dev: ++ nfc_put_device(dev); ++ return rc; + } + + static int nfc_genl_vendor_cmd(struct sk_buff *skb, +@@ -1559,14 +1572,21 @@ static int nfc_genl_vendor_cmd(struct sk_buff *skb, + subcmd = nla_get_u32(info->attrs[NFC_ATTR_VENDOR_SUBCMD]); + + dev = nfc_get_device(dev_idx); +- if (!dev || !dev->vendor_cmds || !dev->n_vendor_cmds) ++ if (!dev) + return -ENODEV; + ++ if (!dev->vendor_cmds || !dev->n_vendor_cmds) { ++ err = -ENODEV; ++ goto put_dev; ++ } ++ + if (info->attrs[NFC_ATTR_VENDOR_DATA]) { + data = nla_data(info->attrs[NFC_ATTR_VENDOR_DATA]); + data_len = nla_len(info->attrs[NFC_ATTR_VENDOR_DATA]); +- if (data_len == 0) +- return -EINVAL; ++ if (data_len == 0) { ++ err = -EINVAL; ++ goto put_dev; ++ } + } else { + data = NULL; + data_len = 0; +@@ -1581,10 +1601,14 @@ static int nfc_genl_vendor_cmd(struct sk_buff *skb, + dev->cur_cmd_info = info; + err = cmd->doit(dev, data, data_len); + dev->cur_cmd_info = NULL; +- return err; ++ goto put_dev; + } + +- return -EOPNOTSUPP; ++ err = -EOPNOTSUPP; ++ ++put_dev: ++ nfc_put_device(dev); ++ return err; + } + + /* message building helper */ +-- +2.35.1 + diff --git a/queue-5.4/nfsd-shut-down-the-nfsv4-state-objects-before-the-fi.patch b/queue-5.4/nfsd-shut-down-the-nfsv4-state-objects-before-the-fi.patch new file mode 100644 index 00000000000..f6b76be5aaa --- /dev/null +++ b/queue-5.4/nfsd-shut-down-the-nfsv4-state-objects-before-the-fi.patch @@ -0,0 +1,42 @@ +From 3e807fb175909014c5f772ceaa325526fb72da24 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 22 Dec 2022 09:51:30 -0500 +Subject: nfsd: shut down the NFSv4 state objects before the filecache + +From: Jeff Layton + +[ Upstream commit 789e1e10f214c00ca18fc6610824c5b9876ba5f2 ] + +Currently, we shut down the filecache before trying to clean up the +stateids that depend on it. This leads to the kernel trying to free an +nfsd_file twice, and a refcount overput on the nf_mark. + +Change the shutdown procedure to tear down all of the stateids prior +to shutting down the filecache. + +Reported-and-tested-by: Wang Yugui +Signed-off-by: Jeff Layton +Fixes: 5e113224c17e ("nfsd: nfsd_file cache entries should be per net namespace") +Signed-off-by: Chuck Lever +Signed-off-by: Sasha Levin +--- + fs/nfsd/nfssvc.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c +index d63cdda1782d..70684c7ae94b 100644 +--- a/fs/nfsd/nfssvc.c ++++ b/fs/nfsd/nfssvc.c +@@ -420,8 +420,8 @@ static void nfsd_shutdown_net(struct net *net) + { + struct nfsd_net *nn = net_generic(net, nfsd_net_id); + +- nfsd_file_cache_shutdown_net(net); + nfs4_state_shutdown_net(net); ++ nfsd_file_cache_shutdown_net(net); + if (nn->lockd_up) { + lockd_down(net); + nn->lockd_up = 0; +-- +2.35.1 + diff --git a/queue-5.4/perf-probe-fix-to-get-the-dw_at_decl_file-and-dw_at_.patch b/queue-5.4/perf-probe-fix-to-get-the-dw_at_decl_file-and-dw_at_.patch new file mode 100644 index 00000000000..bd4275df06b --- /dev/null +++ b/queue-5.4/perf-probe-fix-to-get-the-dw_at_decl_file-and-dw_at_.patch @@ -0,0 +1,92 @@ +From 83600433d190f921a3f5c26d1854ef825496ee05 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Sat, 5 Nov 2022 12:01:14 +0900 +Subject: perf probe: Fix to get the DW_AT_decl_file and DW_AT_call_file as + unsinged data + +From: Masami Hiramatsu (Google) + +[ Upstream commit a9dfc46c67b52ad43b8e335e28f4cf8002c67793 ] + +DWARF version 5 standard Sec 2.14 says that + + Any debugging information entry representing the declaration of an object, + module, subprogram or type may have DW_AT_decl_file, DW_AT_decl_line and + DW_AT_decl_column attributes, each of whose value is an unsigned integer + constant. + +So it should be an unsigned integer data. Also, even though the standard +doesn't clearly say the DW_AT_call_file is signed or unsigned, the +elfutils (eu-readelf) interprets it as unsigned integer data and it is +natural to handle it as unsigned integer data as same as DW_AT_decl_file. +This changes the DW_AT_call_file as unsigned integer data too. + +Fixes: 3f4460a28fb2f73d ("perf probe: Filter out redundant inline-instances") +Signed-off-by: Masami Hiramatsu +Acked-by: Namhyung Kim +Cc: Alexander Shishkin +Cc: Ingo Molnar +Cc: Jiri Olsa +Cc: Mark Rutland +Cc: Masami Hiramatsu +Cc: Peter Zijlstra +Cc: stable@vger.kernel.org +Cc: Steven Rostedt (VMware) +Link: https://lore.kernel.org/r/166761727445.480106.3738447577082071942.stgit@devnote3 +Signed-off-by: Arnaldo Carvalho de Melo +Signed-off-by: Sasha Levin +--- + tools/perf/util/dwarf-aux.c | 21 ++++----------------- + 1 file changed, 4 insertions(+), 17 deletions(-) + +diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c +index b51e0ba363b2..f1e2f566ce6f 100644 +--- a/tools/perf/util/dwarf-aux.c ++++ b/tools/perf/util/dwarf-aux.c +@@ -261,19 +261,6 @@ static int die_get_attr_udata(Dwarf_Die *tp_die, unsigned int attr_name, + return 0; + } + +-/* Get attribute and translate it as a sdata */ +-static int die_get_attr_sdata(Dwarf_Die *tp_die, unsigned int attr_name, +- Dwarf_Sword *result) +-{ +- Dwarf_Attribute attr; +- +- if (dwarf_attr_integrate(tp_die, attr_name, &attr) == NULL || +- dwarf_formsdata(&attr, result) != 0) +- return -ENOENT; +- +- return 0; +-} +- + /** + * die_is_signed_type - Check whether a type DIE is signed or not + * @tp_die: a DIE of a type +@@ -397,9 +384,9 @@ int die_get_data_member_location(Dwarf_Die *mb_die, Dwarf_Word *offs) + /* Get the call file index number in CU DIE */ + static int die_get_call_fileno(Dwarf_Die *in_die) + { +- Dwarf_Sword idx; ++ Dwarf_Word idx; + +- if (die_get_attr_sdata(in_die, DW_AT_call_file, &idx) == 0) ++ if (die_get_attr_udata(in_die, DW_AT_call_file, &idx) == 0) + return (int)idx; + else + return -ENOENT; +@@ -408,9 +395,9 @@ static int die_get_call_fileno(Dwarf_Die *in_die) + /* Get the declared file index number in CU DIE */ + static int die_get_decl_fileno(Dwarf_Die *pdie) + { +- Dwarf_Sword idx; ++ Dwarf_Word idx; + +- if (die_get_attr_sdata(pdie, DW_AT_decl_file, &idx) == 0) ++ if (die_get_attr_udata(pdie, DW_AT_decl_file, &idx) == 0) + return (int)idx; + else + return -ENOENT; +-- +2.35.1 + diff --git a/queue-5.4/perf-probe-use-dwarf_attr_integrate-as-generic-dwarf.patch b/queue-5.4/perf-probe-use-dwarf_attr_integrate-as-generic-dwarf.patch new file mode 100644 index 00000000000..7cbdba6fa38 --- /dev/null +++ b/queue-5.4/perf-probe-use-dwarf_attr_integrate-as-generic-dwarf.patch @@ -0,0 +1,54 @@ +From 9aa5a2b138cf59fd838a7665d93f41ca4a5d2f39 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 1 Nov 2022 22:48:39 +0900 +Subject: perf probe: Use dwarf_attr_integrate as generic DWARF attr accessor + +From: Masami Hiramatsu (Google) + +[ Upstream commit f828929ab7f0dc3353e4a617f94f297fa8f3dec3 ] + +Use dwarf_attr_integrate() instead of dwarf_attr() for generic attribute +acccessor functions, so that it can find the specified attribute from +abstact origin DIE etc. + +Signed-off-by: Masami Hiramatsu +Acked-by: Namhyung Kim +Cc: Alexander Shishkin +Cc: Ingo Molnar +Cc: Jiri Olsa +Cc: Mark Rutland +Cc: Peter Zijlstra +Cc: Steven Rostedt (VMware) +Link: https://lore.kernel.org/r/166731051988.2100653.13595339994343449770.stgit@devnote3 +Signed-off-by: Arnaldo Carvalho de Melo +Stable-dep-of: a9dfc46c67b5 ("perf probe: Fix to get the DW_AT_decl_file and DW_AT_call_file as unsinged data") +Signed-off-by: Sasha Levin +--- + tools/perf/util/dwarf-aux.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c +index ab34ef2c661f..b51e0ba363b2 100644 +--- a/tools/perf/util/dwarf-aux.c ++++ b/tools/perf/util/dwarf-aux.c +@@ -254,7 +254,7 @@ static int die_get_attr_udata(Dwarf_Die *tp_die, unsigned int attr_name, + { + Dwarf_Attribute attr; + +- if (dwarf_attr(tp_die, attr_name, &attr) == NULL || ++ if (dwarf_attr_integrate(tp_die, attr_name, &attr) == NULL || + dwarf_formudata(&attr, result) != 0) + return -ENOENT; + +@@ -267,7 +267,7 @@ static int die_get_attr_sdata(Dwarf_Die *tp_die, unsigned int attr_name, + { + Dwarf_Attribute attr; + +- if (dwarf_attr(tp_die, attr_name, &attr) == NULL || ++ if (dwarf_attr_integrate(tp_die, attr_name, &attr) == NULL || + dwarf_formsdata(&attr, result) != 0) + return -ENOENT; + +-- +2.35.1 + diff --git a/queue-5.4/perf-tools-fix-resources-leak-in-perf_data__open_dir.patch b/queue-5.4/perf-tools-fix-resources-leak-in-perf_data__open_dir.patch new file mode 100644 index 00000000000..c8e1b8a6bd2 --- /dev/null +++ b/queue-5.4/perf-tools-fix-resources-leak-in-perf_data__open_dir.patch @@ -0,0 +1,52 @@ +From 49eb6ec88a10c5179dd366ae1dddf387b4ca92ed Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 29 Dec 2022 13:09:00 +0400 +Subject: perf tools: Fix resources leak in perf_data__open_dir() + +From: Miaoqian Lin + +[ Upstream commit 0a6564ebd953c4590663c9a3c99a3ea9920ade6f ] + +In perf_data__open_dir(), opendir() opens the directory stream. Add +missing closedir() to release it after use. + +Fixes: eb6176709b235b96 ("perf data: Add perf_data__open_dir_data function") +Reviewed-by: Adrian Hunter +Signed-off-by: Miaoqian Lin +Cc: Alexander Shishkin +Cc: Alexey Bayduraev +Cc: Ingo Molnar +Cc: Jiri Olsa +Cc: Mark Rutland +Cc: Namhyung Kim +Cc: Peter Zijlstra +Link: https://lore.kernel.org/r/20221229090903.1402395-1-linmq006@gmail.com +Signed-off-by: Arnaldo Carvalho de Melo +Signed-off-by: Sasha Levin +--- + tools/perf/util/data.c | 2 ++ + 1 file changed, 2 insertions(+) + +diff --git a/tools/perf/util/data.c b/tools/perf/util/data.c +index 3c874f52f1a2..4da900bdb2f1 100644 +--- a/tools/perf/util/data.c ++++ b/tools/perf/util/data.c +@@ -120,6 +120,7 @@ int perf_data__open_dir(struct perf_data *data) + file->size = st.st_size; + } + ++ closedir(dir); + if (!files) + return -EINVAL; + +@@ -128,6 +129,7 @@ int perf_data__open_dir(struct perf_data *data) + return 0; + + out_err: ++ closedir(dir); + close_dir(files, nr); + return ret; + } +-- +2.35.1 + diff --git a/queue-5.4/qlcnic-prevent-dcb-use-after-free-on-qlcnic_dcb_enab.patch b/queue-5.4/qlcnic-prevent-dcb-use-after-free-on-qlcnic_dcb_enab.patch new file mode 100644 index 00000000000..97441044adc --- /dev/null +++ b/queue-5.4/qlcnic-prevent-dcb-use-after-free-on-qlcnic_dcb_enab.patch @@ -0,0 +1,103 @@ +From 448169ea8aa8e1dad9c045d811ce88ba3c3bcbfa Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Thu, 22 Dec 2022 14:52:28 +0300 +Subject: qlcnic: prevent ->dcb use-after-free on qlcnic_dcb_enable() failure + +From: Daniil Tatianin + +[ Upstream commit 13a7c8964afcd8ca43c0b6001ebb0127baa95362 ] + +adapter->dcb would get silently freed inside qlcnic_dcb_enable() in +case qlcnic_dcb_attach() would return an error, which always happens +under OOM conditions. This would lead to use-after-free because both +of the existing callers invoke qlcnic_dcb_get_info() on the obtained +pointer, which is potentially freed at that point. + +Propagate errors from qlcnic_dcb_enable(), and instead free the dcb +pointer at callsite using qlcnic_dcb_free(). This also removes the now +unused qlcnic_clear_dcb_ops() helper, which was a simple wrapper around +kfree() also causing memory leaks for partially initialized dcb. + +Found by Linux Verification Center (linuxtesting.org) with the SVACE +static analysis tool. + +Fixes: 3c44bba1d270 ("qlcnic: Disable DCB operations from SR-IOV VFs") +Reviewed-by: Michal Swiatkowski +Signed-off-by: Daniil Tatianin +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c | 8 +++++++- + drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h | 10 ++-------- + drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c | 8 +++++++- + 3 files changed, 16 insertions(+), 10 deletions(-) + +diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c +index 10286215092f..85419b8258b5 100644 +--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c ++++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_init.c +@@ -2525,7 +2525,13 @@ int qlcnic_83xx_init(struct qlcnic_adapter *adapter, int pci_using_dac) + goto disable_mbx_intr; + + qlcnic_83xx_clear_function_resources(adapter); +- qlcnic_dcb_enable(adapter->dcb); ++ ++ err = qlcnic_dcb_enable(adapter->dcb); ++ if (err) { ++ qlcnic_dcb_free(adapter->dcb); ++ goto disable_mbx_intr; ++ } ++ + qlcnic_83xx_initialize_nic(adapter, 1); + qlcnic_dcb_get_info(adapter->dcb); + +diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h +index 0a9d24e86715..eb8000d9b6d0 100644 +--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h ++++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_dcb.h +@@ -42,11 +42,6 @@ struct qlcnic_dcb { + unsigned long state; + }; + +-static inline void qlcnic_clear_dcb_ops(struct qlcnic_dcb *dcb) +-{ +- kfree(dcb); +-} +- + static inline int qlcnic_dcb_get_hw_capability(struct qlcnic_dcb *dcb) + { + if (dcb && dcb->ops->get_hw_capability) +@@ -113,9 +108,8 @@ static inline void qlcnic_dcb_init_dcbnl_ops(struct qlcnic_dcb *dcb) + dcb->ops->init_dcbnl_ops(dcb); + } + +-static inline void qlcnic_dcb_enable(struct qlcnic_dcb *dcb) ++static inline int qlcnic_dcb_enable(struct qlcnic_dcb *dcb) + { +- if (dcb && qlcnic_dcb_attach(dcb)) +- qlcnic_clear_dcb_ops(dcb); ++ return dcb ? qlcnic_dcb_attach(dcb) : 0; + } + #endif +diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +index 3a96fd6deef7..9d5b74c804b5 100644 +--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c ++++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c +@@ -2639,7 +2639,13 @@ qlcnic_probe(struct pci_dev *pdev, const struct pci_device_id *ent) + "Device does not support MSI interrupts\n"); + + if (qlcnic_82xx_check(adapter)) { +- qlcnic_dcb_enable(adapter->dcb); ++ err = qlcnic_dcb_enable(adapter->dcb); ++ if (err) { ++ qlcnic_dcb_free(adapter->dcb); ++ dev_err(&pdev->dev, "Failed to enable DCB\n"); ++ goto err_out_free_hw; ++ } ++ + qlcnic_dcb_get_info(adapter->dcb); + err = qlcnic_setup_intr(adapter); + +-- +2.35.1 + diff --git a/queue-5.4/ravb-fix-failed-to-switch-device-to-config-mode-mess.patch b/queue-5.4/ravb-fix-failed-to-switch-device-to-config-mode-mess.patch new file mode 100644 index 00000000000..d0a48c401e2 --- /dev/null +++ b/queue-5.4/ravb-fix-failed-to-switch-device-to-config-mode-mess.patch @@ -0,0 +1,68 @@ +From 21f1ad85a9d28639bf41f0013df8e71652969f57 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 14 Dec 2022 10:51:18 +0000 +Subject: ravb: Fix "failed to switch device to config mode" message during + unbind + +From: Biju Das + +[ Upstream commit c72a7e42592b2e18d862cf120876070947000d7a ] + +This patch fixes the error "ravb 11c20000.ethernet eth0: failed to switch +device to config mode" during unbind. + +We are doing register access after pm_runtime_put_sync(). + +We usually do cleanup in reverse order of init. Currently in +remove(), the "pm_runtime_put_sync" is not in reverse order. + +Probe + reset_control_deassert(rstc); + pm_runtime_enable(&pdev->dev); + pm_runtime_get_sync(&pdev->dev); + +remove + pm_runtime_put_sync(&pdev->dev); + unregister_netdev(ndev); + .. + ravb_mdio_release(priv); + pm_runtime_disable(&pdev->dev); + +Consider the call to unregister_netdev() +unregister_netdev->unregister_netdevice_queue->rollback_registered_many +that calls the below functions which access the registers after +pm_runtime_put_sync() + 1) ravb_get_stats + 2) ravb_close + +Fixes: c156633f1353 ("Renesas Ethernet AVB driver proper") +Cc: stable@vger.kernel.org +Signed-off-by: Biju Das +Reviewed-by: Leon Romanovsky +Link: https://lore.kernel.org/r/20221214105118.2495313-1-biju.das.jz@bp.renesas.com +Signed-off-by: Paolo Abeni +Signed-off-by: Sasha Levin +--- + drivers/net/ethernet/renesas/ravb_main.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c +index 95fd1f2d5439..3fd5155bdd5f 100644 +--- a/drivers/net/ethernet/renesas/ravb_main.c ++++ b/drivers/net/ethernet/renesas/ravb_main.c +@@ -2216,11 +2216,11 @@ static int ravb_remove(struct platform_device *pdev) + priv->desc_bat_dma); + /* Set reset mode */ + ravb_write(ndev, CCC_OPC_RESET, CCC); +- pm_runtime_put_sync(&pdev->dev); + unregister_netdev(ndev); + netif_napi_del(&priv->napi[RAVB_NC]); + netif_napi_del(&priv->napi[RAVB_BE]); + ravb_mdio_release(priv); ++ pm_runtime_put_sync(&pdev->dev); + pm_runtime_disable(&pdev->dev); + free_netdev(ndev); + platform_set_drvdata(pdev, NULL); +-- +2.35.1 + diff --git a/queue-5.4/rdma-mlx5-fix-validation-of-max_rd_atomic-caps-for-d.patch b/queue-5.4/rdma-mlx5-fix-validation-of-max_rd_atomic-caps-for-d.patch new file mode 100644 index 00000000000..d3ed5376912 --- /dev/null +++ b/queue-5.4/rdma-mlx5-fix-validation-of-max_rd_atomic-caps-for-d.patch @@ -0,0 +1,95 @@ +From 18b1fdf952b0e3cb095182f80a241e69bd4d9773 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 28 Dec 2022 14:56:10 +0200 +Subject: RDMA/mlx5: Fix validation of max_rd_atomic caps for DC + +From: Maor Gottlieb + +[ Upstream commit 8de8482fe5732fbef4f5af82bc0c0362c804cd1f ] + +Currently, when modifying DC, we validate max_rd_atomic user attribute +against the RC cap, validate against DC. RC and DC QP types have different +device limitations. + +This can cause userspace created DC QPs to malfunction. + +Fixes: c32a4f296e1d ("IB/mlx5: Add support for DC Initiator QP") +Link: https://lore.kernel.org/r/0c5aee72cea188c3bb770f4207cce7abc9b6fc74.1672231736.git.leonro@nvidia.com +Signed-off-by: Maor Gottlieb +Signed-off-by: Leon Romanovsky +Signed-off-by: Sasha Levin +--- + drivers/infiniband/hw/mlx5/qp.c | 49 +++++++++++++++++++++++---------- + 1 file changed, 35 insertions(+), 14 deletions(-) + +diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c +index 634f29cb7395..6edd30c92156 100644 +--- a/drivers/infiniband/hw/mlx5/qp.c ++++ b/drivers/infiniband/hw/mlx5/qp.c +@@ -3890,6 +3890,40 @@ static int mlx5_ib_modify_dct(struct ib_qp *ibqp, struct ib_qp_attr *attr, + return err; + } + ++static int validate_rd_atomic(struct mlx5_ib_dev *dev, struct ib_qp_attr *attr, ++ int attr_mask, enum ib_qp_type qp_type) ++{ ++ int log_max_ra_res; ++ int log_max_ra_req; ++ ++ if (qp_type == MLX5_IB_QPT_DCI) { ++ log_max_ra_res = 1 << MLX5_CAP_GEN(dev->mdev, ++ log_max_ra_res_dc); ++ log_max_ra_req = 1 << MLX5_CAP_GEN(dev->mdev, ++ log_max_ra_req_dc); ++ } else { ++ log_max_ra_res = 1 << MLX5_CAP_GEN(dev->mdev, ++ log_max_ra_res_qp); ++ log_max_ra_req = 1 << MLX5_CAP_GEN(dev->mdev, ++ log_max_ra_req_qp); ++ } ++ ++ if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC && ++ attr->max_rd_atomic > log_max_ra_res) { ++ mlx5_ib_dbg(dev, "invalid max_rd_atomic value %d\n", ++ attr->max_rd_atomic); ++ return false; ++ } ++ ++ if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC && ++ attr->max_dest_rd_atomic > log_max_ra_req) { ++ mlx5_ib_dbg(dev, "invalid max_dest_rd_atomic value %d\n", ++ attr->max_dest_rd_atomic); ++ return false; ++ } ++ return true; ++} ++ + int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, + int attr_mask, struct ib_udata *udata) + { +@@ -3986,21 +4020,8 @@ int mlx5_ib_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, + } + } + +- if (attr_mask & IB_QP_MAX_QP_RD_ATOMIC && +- attr->max_rd_atomic > +- (1 << MLX5_CAP_GEN(dev->mdev, log_max_ra_res_qp))) { +- mlx5_ib_dbg(dev, "invalid max_rd_atomic value %d\n", +- attr->max_rd_atomic); +- goto out; +- } +- +- if (attr_mask & IB_QP_MAX_DEST_RD_ATOMIC && +- attr->max_dest_rd_atomic > +- (1 << MLX5_CAP_GEN(dev->mdev, log_max_ra_req_qp))) { +- mlx5_ib_dbg(dev, "invalid max_dest_rd_atomic value %d\n", +- attr->max_dest_rd_atomic); ++ if (!validate_rd_atomic(dev, attr, attr_mask, qp_type)) + goto out; +- } + + if (cur_state == new_state && cur_state == IB_QPS_RESET) { + err = 0; +-- +2.35.1 + diff --git a/queue-5.4/rdma-uverbs-silence-shifttoomanybitssigned-warning.patch b/queue-5.4/rdma-uverbs-silence-shifttoomanybitssigned-warning.patch new file mode 100644 index 00000000000..b5aee3f0398 --- /dev/null +++ b/queue-5.4/rdma-uverbs-silence-shifttoomanybitssigned-warning.patch @@ -0,0 +1,40 @@ +From 7015fff14a0ab737684dc4af1e6793aafbbf8a29 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 20 Jul 2020 20:56:27 +0300 +Subject: RDMA/uverbs: Silence shiftTooManyBitsSigned warning + +From: Leon Romanovsky + +[ Upstream commit 9b8d846924856570625b93f83ae0624391193bce ] + +Fix reported by kbuild warning. + + drivers/infiniband/core/uverbs_cmd.c:1897:47: warning: Shifting signed 32-bit value by 31 bits is undefined behaviour [shiftTooManyBitsSigned] + BUILD_BUG_ON(IB_USER_LAST_QP_ATTR_MASK == (1 << 31)); + ^ +Link: https://lore.kernel.org/r/20200720175627.1273096-3-leon@kernel.org +Reported-by: kernel test robot +Signed-off-by: Leon Romanovsky +Signed-off-by: Jason Gunthorpe +Stable-dep-of: 8de8482fe573 ("RDMA/mlx5: Fix validation of max_rd_atomic caps for DC") +Signed-off-by: Sasha Levin +--- + drivers/infiniband/core/uverbs_cmd.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c +index d413dafb9211..39cbb853f913 100644 +--- a/drivers/infiniband/core/uverbs_cmd.c ++++ b/drivers/infiniband/core/uverbs_cmd.c +@@ -1952,7 +1952,7 @@ static int ib_uverbs_ex_modify_qp(struct uverbs_attr_bundle *attrs) + * Last bit is reserved for extending the attr_mask by + * using another field. + */ +- BUILD_BUG_ON(IB_USER_LAST_QP_ATTR_MASK == (1 << 31)); ++ BUILD_BUG_ON(IB_USER_LAST_QP_ATTR_MASK == (1ULL << 31)); + + if (cmd.base.attr_mask & + ~((IB_USER_LAST_QP_ATTR_MASK << 1) - 1)) +-- +2.35.1 + diff --git a/queue-5.4/riscv-stacktrace-fix-stack-output-without-ra-on-the-.patch b/queue-5.4/riscv-stacktrace-fix-stack-output-without-ra-on-the-.patch new file mode 100644 index 00000000000..966c534d6f7 --- /dev/null +++ b/queue-5.4/riscv-stacktrace-fix-stack-output-without-ra-on-the-.patch @@ -0,0 +1,72 @@ +From e92954d051587a06c852db0cb6f3173e9f9f651e Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Mon, 11 Jan 2021 20:40:14 +0800 +Subject: riscv/stacktrace: Fix stack output without ra on the stack top + +From: Chen Huang + +[ Upstream commit f766f77a74f5784d8d4d3c36b1900731f97d08d0 ] + +When a function doesn't have a callee, then it will not +push ra into the stack, such as lkdtm_BUG() function, + +addi sp,sp,-16 +sd s0,8(sp) +addi s0,sp,16 +ebreak + +The struct stackframe use {fp,ra} to get information from +stack, if walk_stackframe() with pr_regs, we will obtain +wrong value and bad stacktrace, + +[] lkdtm_BUG+0x6/0x8 +---[ end trace 18da3fbdf08e25d5 ]--- + +Correct the next fp and pc, after that, full stacktrace +shown as expects, + +[] lkdtm_BUG+0x6/0x8 +[] lkdtm_do_action+0x14/0x1c +[] direct_entry+0xc0/0x10a +[] full_proxy_write+0x42/0x6a +[] vfs_write+0x7e/0x214 +[] ksys_write+0x98/0xc0 +[] sys_write+0xe/0x16 +[] ret_from_syscall+0x0/0x2 +---[ end trace 61917f3d9a9fadcd ]--- + +Signed-off-by: Chen Huang +Signed-off-by: Kefeng Wang +Signed-off-by: Palmer Dabbelt +Stable-dep-of: 5c3022e4a616 ("riscv: stacktrace: Fixup ftrace_graph_ret_addr retp argument") +Signed-off-by: Sasha Levin +--- + arch/riscv/kernel/stacktrace.c | 12 +++++++++--- + 1 file changed, 9 insertions(+), 3 deletions(-) + +diff --git a/arch/riscv/kernel/stacktrace.c b/arch/riscv/kernel/stacktrace.c +index 19e46f4160cc..1a512a24879e 100644 +--- a/arch/riscv/kernel/stacktrace.c ++++ b/arch/riscv/kernel/stacktrace.c +@@ -55,9 +55,15 @@ void notrace walk_stackframe(struct task_struct *task, struct pt_regs *regs, + /* Unwind stack frame */ + frame = (struct stackframe *)fp - 1; + sp = fp; +- fp = frame->fp; +- pc = ftrace_graph_ret_addr(current, NULL, frame->ra, +- (unsigned long *)(fp - 8)); ++ if (regs && (regs->epc == pc) && (frame->fp & 0x7)) { ++ fp = frame->ra; ++ pc = regs->ra; ++ } else { ++ fp = frame->fp; ++ pc = ftrace_graph_ret_addr(current, NULL, frame->ra, ++ (unsigned long *)(fp - 8)); ++ } ++ + } + } + +-- +2.35.1 + diff --git a/queue-5.4/riscv-stacktrace-fixup-ftrace_graph_ret_addr-retp-ar.patch b/queue-5.4/riscv-stacktrace-fixup-ftrace_graph_ret_addr-retp-ar.patch new file mode 100644 index 00000000000..43f712c7231 --- /dev/null +++ b/queue-5.4/riscv-stacktrace-fixup-ftrace_graph_ret_addr-retp-ar.patch @@ -0,0 +1,41 @@ +From eacfe5acf0bf82c6079fc7250a2c8731b481e7a6 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 9 Nov 2022 01:49:36 -0500 +Subject: riscv: stacktrace: Fixup ftrace_graph_ret_addr retp argument + +From: Guo Ren + +[ Upstream commit 5c3022e4a616d800cf5f4c3a981d7992179e44a1 ] + +The 'retp' is a pointer to the return address on the stack, so we +must pass the current return address pointer as the 'retp' +argument to ftrace_push_return_trace(). Not parent function's +return address on the stack. + +Fixes: b785ec129bd9 ("riscv/ftrace: Add HAVE_FUNCTION_GRAPH_RET_ADDR_PTR support") +Signed-off-by: Guo Ren +Signed-off-by: Guo Ren +Link: https://lore.kernel.org/r/20221109064937.3643993-2-guoren@kernel.org +Cc: stable@vger.kernel.org +Signed-off-by: Palmer Dabbelt +Signed-off-by: Sasha Levin +--- + arch/riscv/kernel/stacktrace.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/arch/riscv/kernel/stacktrace.c b/arch/riscv/kernel/stacktrace.c +index 1a512a24879e..a1ee7f33c205 100644 +--- a/arch/riscv/kernel/stacktrace.c ++++ b/arch/riscv/kernel/stacktrace.c +@@ -61,7 +61,7 @@ void notrace walk_stackframe(struct task_struct *task, struct pt_regs *regs, + } else { + fp = frame->fp; + pc = ftrace_graph_ret_addr(current, NULL, frame->ra, +- (unsigned long *)(fp - 8)); ++ &frame->ra); + } + + } +-- +2.35.1 + diff --git a/queue-5.4/series b/queue-5.4/series index a257910ab0d..7dd9ed5ee6c 100644 --- a/queue-5.4/series +++ b/queue-5.4/series @@ -540,3 +540,51 @@ pm-devfreq-governor-add-a-private-governor_data-for-.patch media-s5p-mfc-fix-to-handle-reference-queue-during-f.patch media-s5p-mfc-clear-workbit-to-handle-error-conditio.patch media-s5p-mfc-fix-in-register-read-and-write-for-h26.patch +dm-thin-resume-even-if-in-fail-mode.patch +perf-probe-use-dwarf_attr_integrate-as-generic-dwarf.patch +perf-probe-fix-to-get-the-dw_at_decl_file-and-dw_at_.patch +kvm-x86-optimize-more-exit-handlers-in-vmx.c.patch +kvm-retpolines-x86-eliminate-retpoline-from-vmx.c-ex.patch +kvm-vmx-rename-interrupt_pending-to-interrupt_window.patch +kvm-vmx-rename-nmi_pending-to-nmi_window.patch +kvm-vmx-fix-the-spelling-of-cpu_based_use_tsc_offset.patch +kvm-nvmx-properly-expose-enable_usr_wait_pause-contr.patch +ravb-fix-failed-to-switch-device-to-config-mode-mess.patch +riscv-stacktrace-fix-stack-output-without-ra-on-the-.patch +riscv-stacktrace-fixup-ftrace_graph_ret_addr-retp-ar.patch +driver-core-fix-driver_deferred_probe_check_state-lo.patch +driver-core-set-deferred_probe_timeout-to-a-longer-d.patch +ext4-goto-right-label-failed_mount3a.patch +ext4-correct-inconsistent-error-msg-in-nojournal-mod.patch +mm-highmem-lift-memcpy_-to-from-_page-to-core.patch +ext4-use-memcpy_to_page-in-pagecache_write.patch +fs-ext4-initialize-fsdata-in-pagecache_write.patch +ext4-use-kmemdup-to-replace-kmalloc-memcpy.patch +mbcache-don-t-reclaim-used-entries.patch +mbcache-add-functions-to-delete-entry-if-unused.patch +ext4-remove-ea-inode-entry-from-mbcache-on-inode-evi.patch +ext4-unindent-codeblock-in-ext4_xattr_block_set.patch +ext4-fix-race-when-reusing-xattr-blocks.patch +mbcache-automatically-delete-entries-from-cache-on-f.patch +ext4-fix-deadlock-due-to-mbcache-entry-corruption.patch +sunrpc-ensure-the-matching-upcall-is-in-flight-upon-.patch +bpf-pull-before-calling-skb_postpull_rcsum.patch +nfsd-shut-down-the-nfsv4-state-objects-before-the-fi.patch +net-hns3-add-interrupts-re-initialization-while-doin.patch +net-sched-fix-memory-leak-in-tcindex_set_parms.patch +qlcnic-prevent-dcb-use-after-free-on-qlcnic_dcb_enab.patch +nfc-fix-potential-resource-leaks.patch +vhost-fix-range-used-in-translate_desc.patch +net-amd-xgbe-add-missed-tasklet_kill.patch +net-phy-xgmiitorgmii-fix-refcount-leak-in-xgmiitorgm.patch +rdma-uverbs-silence-shifttoomanybitssigned-warning.patch +rdma-mlx5-fix-validation-of-max_rd_atomic-caps-for-d.patch +net-sched-atm-dont-intepret-cls-results-when-asked-t.patch +net-sched-cbq-dont-intepret-cls-results-when-asked-t.patch +perf-tools-fix-resources-leak-in-perf_data__open_dir.patch +drivers-net-bonding-bond_3ad-return-when-there-s-no-.patch +usb-rndis_host-secure-rndis_query-check-against-int-.patch +drm-i915-unpin-on-error-in-intel_vgpu_shadow_mm_pin.patch +caif-fix-memory-leak-in-cfctrl_linkup_request.patch +udf-fix-extension-of-the-last-extent-in-the-file.patch +asoc-intel-bytcr_rt5640-add-quirk-for-the-advantech-.patch diff --git a/queue-5.4/sunrpc-ensure-the-matching-upcall-is-in-flight-upon-.patch b/queue-5.4/sunrpc-ensure-the-matching-upcall-is-in-flight-upon-.patch new file mode 100644 index 00000000000..f4e35b990dc --- /dev/null +++ b/queue-5.4/sunrpc-ensure-the-matching-upcall-is-in-flight-upon-.patch @@ -0,0 +1,133 @@ +From e23beb280a5642b830a0c1c0657730f9158bd0de Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 13 Dec 2022 13:14:31 +0900 +Subject: SUNRPC: ensure the matching upcall is in-flight upon downcall + +From: minoura makoto + +[ Upstream commit b18cba09e374637a0a3759d856a6bca94c133952 ] + +Commit 9130b8dbc6ac ("SUNRPC: allow for upcalls for the same uid +but different gss service") introduced `auth` argument to +__gss_find_upcall(), but in gss_pipe_downcall() it was left as NULL +since it (and auth->service) was not (yet) determined. + +When multiple upcalls with the same uid and different service are +ongoing, it could happen that __gss_find_upcall(), which returns the +first match found in the pipe->in_downcall list, could not find the +correct gss_msg corresponding to the downcall we are looking for. +Moreover, it might return a msg which is not sent to rpc.gssd yet. + +We could see mount.nfs process hung in D state with multiple mount.nfs +are executed in parallel. The call trace below is of CentOS 7.9 +kernel-3.10.0-1160.24.1.el7.x86_64 but we observed the same hang w/ +elrepo kernel-ml-6.0.7-1.el7. + +PID: 71258 TASK: ffff91ebd4be0000 CPU: 36 COMMAND: "mount.nfs" + #0 [ffff9203ca3234f8] __schedule at ffffffffa3b8899f + #1 [ffff9203ca323580] schedule at ffffffffa3b88eb9 + #2 [ffff9203ca323590] gss_cred_init at ffffffffc0355818 [auth_rpcgss] + #3 [ffff9203ca323658] rpcauth_lookup_credcache at ffffffffc0421ebc +[sunrpc] + #4 [ffff9203ca3236d8] gss_lookup_cred at ffffffffc0353633 [auth_rpcgss] + #5 [ffff9203ca3236e8] rpcauth_lookupcred at ffffffffc0421581 [sunrpc] + #6 [ffff9203ca323740] rpcauth_refreshcred at ffffffffc04223d3 [sunrpc] + #7 [ffff9203ca3237a0] call_refresh at ffffffffc04103dc [sunrpc] + #8 [ffff9203ca3237b8] __rpc_execute at ffffffffc041e1c9 [sunrpc] + #9 [ffff9203ca323820] rpc_execute at ffffffffc0420a48 [sunrpc] + +The scenario is like this. Let's say there are two upcalls for +services A and B, A -> B in pipe->in_downcall, B -> A in pipe->pipe. + +When rpc.gssd reads pipe to get the upcall msg corresponding to +service B from pipe->pipe and then writes the response, in +gss_pipe_downcall the msg corresponding to service A will be picked +because only uid is used to find the msg and it is before the one for +B in pipe->in_downcall. And the process waiting for the msg +corresponding to service A will be woken up. + +Actual scheduing of that process might be after rpc.gssd processes the +next msg. In rpc_pipe_generic_upcall it clears msg->errno (for A). +The process is scheduled to see gss_msg->ctx == NULL and +gss_msg->msg.errno == 0, therefore it cannot break the loop in +gss_create_upcall and is never woken up after that. + +This patch adds a simple check to ensure that a msg which is not +sent to rpc.gssd yet is not chosen as the matching upcall upon +receiving a downcall. + +Signed-off-by: minoura makoto +Signed-off-by: Hiroshi Shimamoto +Tested-by: Hiroshi Shimamoto +Cc: Trond Myklebust +Fixes: 9130b8dbc6ac ("SUNRPC: allow for upcalls for same uid but different gss service") +Signed-off-by: Trond Myklebust +Signed-off-by: Sasha Levin +--- + include/linux/sunrpc/rpc_pipe_fs.h | 5 +++++ + net/sunrpc/auth_gss/auth_gss.c | 19 +++++++++++++++++-- + 2 files changed, 22 insertions(+), 2 deletions(-) + +diff --git a/include/linux/sunrpc/rpc_pipe_fs.h b/include/linux/sunrpc/rpc_pipe_fs.h +index e90b9bd99ded..396de2ef8767 100644 +--- a/include/linux/sunrpc/rpc_pipe_fs.h ++++ b/include/linux/sunrpc/rpc_pipe_fs.h +@@ -94,6 +94,11 @@ extern ssize_t rpc_pipe_generic_upcall(struct file *, struct rpc_pipe_msg *, + char __user *, size_t); + extern int rpc_queue_upcall(struct rpc_pipe *, struct rpc_pipe_msg *); + ++/* returns true if the msg is in-flight, i.e., already eaten by the peer */ ++static inline bool rpc_msg_is_inflight(const struct rpc_pipe_msg *msg) { ++ return (msg->copied != 0 && list_empty(&msg->list)); ++} ++ + struct rpc_clnt; + extern struct dentry *rpc_create_client_dir(struct dentry *, const char *, struct rpc_clnt *); + extern int rpc_remove_client_dir(struct rpc_clnt *); +diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c +index b7a71578bd98..4d3cf146f50a 100644 +--- a/net/sunrpc/auth_gss/auth_gss.c ++++ b/net/sunrpc/auth_gss/auth_gss.c +@@ -301,7 +301,7 @@ __gss_find_upcall(struct rpc_pipe *pipe, kuid_t uid, const struct gss_auth *auth + list_for_each_entry(pos, &pipe->in_downcall, list) { + if (!uid_eq(pos->uid, uid)) + continue; +- if (auth && pos->auth->service != auth->service) ++ if (pos->auth->service != auth->service) + continue; + refcount_inc(&pos->count); + return pos; +@@ -683,6 +683,21 @@ gss_create_upcall(struct gss_auth *gss_auth, struct gss_cred *gss_cred) + return err; + } + ++static struct gss_upcall_msg * ++gss_find_downcall(struct rpc_pipe *pipe, kuid_t uid) ++{ ++ struct gss_upcall_msg *pos; ++ list_for_each_entry(pos, &pipe->in_downcall, list) { ++ if (!uid_eq(pos->uid, uid)) ++ continue; ++ if (!rpc_msg_is_inflight(&pos->msg)) ++ continue; ++ refcount_inc(&pos->count); ++ return pos; ++ } ++ return NULL; ++} ++ + #define MSG_BUF_MAXSIZE 1024 + + static ssize_t +@@ -729,7 +744,7 @@ gss_pipe_downcall(struct file *filp, const char __user *src, size_t mlen) + err = -ENOENT; + /* Find a matching upcall */ + spin_lock(&pipe->lock); +- gss_msg = __gss_find_upcall(pipe, uid, NULL); ++ gss_msg = gss_find_downcall(pipe, uid); + if (gss_msg == NULL) { + spin_unlock(&pipe->lock); + goto err_put_ctx; +-- +2.35.1 + diff --git a/queue-5.4/udf-fix-extension-of-the-last-extent-in-the-file.patch b/queue-5.4/udf-fix-extension-of-the-last-extent-in-the-file.patch new file mode 100644 index 00000000000..7f501995bbf --- /dev/null +++ b/queue-5.4/udf-fix-extension-of-the-last-extent-in-the-file.patch @@ -0,0 +1,37 @@ +From 6c5a87ab8ba62c5768392044c4c0101d1390ee5f Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 21 Dec 2022 17:45:51 +0100 +Subject: udf: Fix extension of the last extent in the file + +From: Jan Kara + +[ Upstream commit 83c7423d1eb6806d13c521d1002cc1a012111719 ] + +When extending the last extent in the file within the last block, we +wrongly computed the length of the last extent. This is mostly a +cosmetical problem since the extent does not contain any data and the +length will be fixed up by following operations but still. + +Fixes: 1f3868f06855 ("udf: Fix extending file within last block") +Signed-off-by: Jan Kara +Signed-off-by: Sasha Levin +--- + fs/udf/inode.c | 2 +- + 1 file changed, 1 insertion(+), 1 deletion(-) + +diff --git a/fs/udf/inode.c b/fs/udf/inode.c +index f6bbf395ce07..37a6bbd5a19c 100644 +--- a/fs/udf/inode.c ++++ b/fs/udf/inode.c +@@ -602,7 +602,7 @@ static void udf_do_extend_final_block(struct inode *inode, + */ + if (new_elen <= (last_ext->extLength & UDF_EXTENT_LENGTH_MASK)) + return; +- added_bytes = (last_ext->extLength & UDF_EXTENT_LENGTH_MASK) - new_elen; ++ added_bytes = new_elen - (last_ext->extLength & UDF_EXTENT_LENGTH_MASK); + last_ext->extLength += added_bytes; + UDF_I(inode)->i_lenExtents += added_bytes; + +-- +2.35.1 + diff --git a/queue-5.4/usb-rndis_host-secure-rndis_query-check-against-int-.patch b/queue-5.4/usb-rndis_host-secure-rndis_query-check-against-int-.patch new file mode 100644 index 00000000000..09b11041e26 --- /dev/null +++ b/queue-5.4/usb-rndis_host-secure-rndis_query-check-against-int-.patch @@ -0,0 +1,43 @@ +From 2864f82d070bdf818bc8ee3f022e224f8f2f1540 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Tue, 3 Jan 2023 10:17:09 +0100 +Subject: usb: rndis_host: Secure rndis_query check against int overflow + +From: Szymon Heidrich + +[ Upstream commit c7dd13805f8b8fc1ce3b6d40f6aff47e66b72ad2 ] + +Variables off and len typed as uint32 in rndis_query function +are controlled by incoming RNDIS response message thus their +value may be manipulated. Setting off to a unexpectetly large +value will cause the sum with len and 8 to overflow and pass +the implemented validation step. Consequently the response +pointer will be referring to a location past the expected +buffer boundaries allowing information leakage e.g. via +RNDIS_OID_802_3_PERMANENT_ADDRESS OID. + +Fixes: ddda08624013 ("USB: rndis_host, various cleanups") +Signed-off-by: Szymon Heidrich +Signed-off-by: David S. Miller +Signed-off-by: Sasha Levin +--- + drivers/net/usb/rndis_host.c | 3 ++- + 1 file changed, 2 insertions(+), 1 deletion(-) + +diff --git a/drivers/net/usb/rndis_host.c b/drivers/net/usb/rndis_host.c +index 1505fe3f87ed..1ff723e15d52 100644 +--- a/drivers/net/usb/rndis_host.c ++++ b/drivers/net/usb/rndis_host.c +@@ -255,7 +255,8 @@ static int rndis_query(struct usbnet *dev, struct usb_interface *intf, + + off = le32_to_cpu(u.get_c->offset); + len = le32_to_cpu(u.get_c->len); +- if (unlikely((8 + off + len) > CONTROL_BUFFER_SIZE)) ++ if (unlikely((off > CONTROL_BUFFER_SIZE - 8) || ++ (len > CONTROL_BUFFER_SIZE - 8 - off))) + goto response_error; + + if (*reply_len != -1 && len != *reply_len) +-- +2.35.1 + diff --git a/queue-5.4/vhost-fix-range-used-in-translate_desc.patch b/queue-5.4/vhost-fix-range-used-in-translate_desc.patch new file mode 100644 index 00000000000..48167482035 --- /dev/null +++ b/queue-5.4/vhost-fix-range-used-in-translate_desc.patch @@ -0,0 +1,55 @@ +From 09be1b540fae534164adeefb63cf9f0e419e02c6 Mon Sep 17 00:00:00 2001 +From: Sasha Levin +Date: Wed, 9 Nov 2022 11:25:03 +0100 +Subject: vhost: fix range used in translate_desc() + +From: Stefano Garzarella + +[ Upstream commit 98047313cdb46828093894d0ac8b1183b8b317f9 ] + +vhost_iotlb_itree_first() requires `start` and `last` parameters +to search for a mapping that overlaps the range. + +In translate_desc() we cyclically call vhost_iotlb_itree_first(), +incrementing `addr` by the amount already translated, so rightly +we move the `start` parameter passed to vhost_iotlb_itree_first(), +but we should hold the `last` parameter constant. + +Let's fix it by saving the `last` parameter value before incrementing +`addr` in the loop. + +Fixes: a9709d6874d5 ("vhost: convert pre sorted vhost memory array to interval tree") +Acked-by: Jason Wang +Signed-off-by: Stefano Garzarella +Message-Id: <20221109102503.18816-3-sgarzare@redhat.com> +Signed-off-by: Michael S. Tsirkin +Signed-off-by: Sasha Levin +--- + drivers/vhost/vhost.c | 4 ++-- + 1 file changed, 2 insertions(+), 2 deletions(-) + +diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c +index 97be299f0a8d..fdfa399700fe 100644 +--- a/drivers/vhost/vhost.c ++++ b/drivers/vhost/vhost.c +@@ -2050,7 +2050,7 @@ static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len, + struct vhost_dev *dev = vq->dev; + struct vhost_umem *umem = dev->iotlb ? dev->iotlb : dev->umem; + struct iovec *_iov; +- u64 s = 0; ++ u64 s = 0, last = addr + len - 1; + int ret = 0; + + while ((u64)len > s) { +@@ -2061,7 +2061,7 @@ static int translate_desc(struct vhost_virtqueue *vq, u64 addr, u32 len, + } + + node = vhost_umem_interval_tree_iter_first(&umem->umem_tree, +- addr, addr + len - 1); ++ addr, last); + if (node == NULL || node->start > addr) { + if (umem != dev->iotlb) { + ret = -EFAULT; +-- +2.35.1 +