ASoC: soc-pcm: use GFP_KERNEL when the code is sleepable
Subsystems (9)
Sasha Levin
maintainerPATCH 1/22 ASoC: soc-pcm: use GFP_KERNEL when the code is sleepable
From: Christophe JAILLET <[email protected]>
[ Upstream commit fb6d679fee95d272c0a94912c4e534146823ee89 ]
At the kzalloc() call in dpcm_be_connect(), there is no spin lock involved. It's merely protected by card->pcm_mutex, instead. The spinlock is applied at the later call with snd_soc_pcm_stream_lock_irq() only for the list manipulations. (See it's *_irq(), not *_irqsave(); that means the context being sleepable at that point.) So, we can use GFP_KERNEL safely there.
This patch revert commit d8a9c6e1f676 ("ASoC: soc-pcm: use GFP_ATOMIC for dpcm structure") which is no longer needed since commit b7898396f4bb ("ASoC: soc-pcm: Fix and cleanup DPCM locking").
--- sound/soc/soc-pcm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/soc-pcm.c +1 -1
@@ -1214,7 +1214,7 @@ static int dpcm_be_connect(struct snd_soc_pcm_runtime *fe, be_substream->pcm->nonatomic =1 ; } - dpcm =kzalloc (sizeof (struct snd_soc_dpcm ),GFP_ATOMIC ); + dpcm =kzalloc (sizeof (struct snd_soc_dpcm ),GFP_KERNEL );if (!dpcm)return -ENOMEM ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 2/22 ASoC: cs35l41: Fix a shift-out-of-bounds warning found by UBSAN
From: Hui Wang <[email protected]>
[ Upstream commit 0b3d5d2e358ca6772fc3662fca27acb12a682fbf ]
We enabled UBSAN in the ubuntu kernel, and the cs35l41 driver triggers a warning calltrace like below:
cs35l41-hda i2c-CSC3551:00-cs35l41-hda.0: bitoffset= 8, word_offset=23, bit_sum mod 32=0, otp_map[i].size = 24 cs35l41-hda i2c-CSC3551:00-cs35l41-hda.0: bitoffset= 0, word_offset=24, bit_sum mod 32=24, otp_map[i].size = 0
================================================================================
UBSAN: shift-out-of-bounds in linux-kernel-src/sound/soc/codecs/cs35l41-lib.c:836:8 shift exponent 64 is too large for 64-bit type 'long unsigned int' CPU: 10 PID: 595 Comm: systemd-udevd Not tainted 5.15.0-23-generic #23 Hardware name: LENOVO \x02MFG_IN_GO/\x02MFG_IN_GO, BIOS N3GET19W (1.00 ) 03/11/2022 Call Trace: <TASK> show_stack+0x52/0x58 dump_stack_lvl+0x4a/0x5f dump_stack+0x10/0x12 ubsan_epilogue+0x9/0x45 __ubsan_handle_shift_out_of_bounds.cold+0x61/0xef ? regmap_unlock_mutex+0xe/0x10 cs35l41_otp_unpack.cold+0x1c6/0x2b2 [snd_soc_cs35l41_lib] cs35l41_hda_probe+0x24f/0x33a [snd_hda_scodec_cs35l41] cs35l41_hda_i2c_probe+0x65/0x90 [snd_hda_scodec_cs35l41_i2c]
When both bitoffset and otp_map[i].size are 0, the line 836 will result in GENMASK(-1, 0), this triggers the shift-out-of-bounds calltrace.
Here add a checking, if both bitoffset and otp_map[i].size are 0, do not run GENMASK() and directly set otp_val to 0, this will not bring any function change on the driver but could avoid the calltrace.
--- sound/soc/codecs/cs35l41-lib.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
sound/soc/codecs/cs35l41-lib.c +4 -2
@@ -831,12 +831,14 @@ int cs35l41_otp_unpack(struct device *dev, struct regmap *regmap) GENMASK(bit_offset + otp_map[i].size - 33, 0)) << (32 - bit_offset); bit_offset += otp_map[i].size -32 ; - }else { + }else if (bit_offset + otp_map[i].size -1 >=0 ) { otp_val = (otp_mem[word_offset] &GENMASK (bit_offset + otp_map[i].size -1 , bit_offset) ) >> bit_offset; bit_offset += otp_map[i].size; - } + }else /* both bit_offset and otp_map[i].size are 0 */ + otp_val =0 ; + bit_sum += otp_map[i].size;if (bit_offset ==32 ) {
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 3/22 ASoC: rt711/5682: check if bus is active before deferred jack detection
In reply to Sasha LevinFrom: Pierre-Louis Bossart <[email protected]>
[ Upstream commit 770f3d992a3f7330f801dfeee98429b2885c9fdb ]
This patch takes a defensive programming and paranoid approach in case the parent device (SoundWire) is pm_runtime resumed but the rt711 device is not. In that case, during the attachment and initialization, a jack detection workqueue can be scheduled. Since the pm_runtime suspend routines will not be invoked, the sequence to cancel all deferred work is not executed, and the jack detection could happen after the bus stops operating, leading to a timeout.
This patch applies the same solution to rt5682, based on the similarities between codec drivers. The race condition with rt5682 was not detected experimentally though.
--- sound/soc/codecs/rt5682.c | 9 +++++++++ sound/soc/codecs/rt711.c | 7 +++++++ 2 files changed, 16 insertions(+)
sound/soc/codecs/rt5682.c +9 -0
@@ -1100,6 +1100,15 @@ void rt5682_jack_detect_handler(struct work_struct *work)return ; } +if (rt5682->is_sdw) { +if (pm_runtime_status_suspended (rt5682->slave->dev.parent)) { +dev_dbg (&rt5682->slave->dev, +"%s: parent device is pm_runtime_status_suspended, skipping jack detection\n" , + __func__); +return ; + } + } + dapm =snd_soc_component_get_dapm (rt5682->component);snd_soc_dapm_mutex_lock (dapm);
sound/soc/codecs/rt711.c +7 -0
@@ -245,6 +245,13 @@ static void rt711_jack_detect_handler(struct work_struct *work)if (!rt711->component->card->instantiated)return ; +if (pm_runtime_status_suspended (rt711->slave->dev.parent)) { +dev_dbg (&rt711->slave->dev, +"%s: parent device is pm_runtime_status_suspended, skipping jack detection\n" , + __func__); +return ; + } + reg =RT711_VERB_GET_PIN_SENSE |RT711_HP_OUT ; ret =regmap_read (rt711->regmap, reg, &jack_status);if (ret <0 )
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 4/22 ASoC: Intel: soc-acpi: correct device endpoints for max98373
In reply to Sasha LevinFrom: Chao Song <[email protected]>
[ Upstream commit 97326be14df7bacc6ba5c62c0556298c27ea0432 ]
The left speaker of max98373 uses spk_r_endpoint, and right speaker uses spk_l_endpoint, this is obviously wrong.
This patch corrects the endpoints for max98373 codec.
--- sound/soc/intel/common/soc-acpi-intel-tgl-match.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
sound/soc/intel/common/soc-acpi-intel-tgl-match.c +2 -2
@@ -132,13 +132,13 @@ static const struct snd_soc_acpi_adr_device mx8373_1_adr[] = { { .adr = 0x000123019F837300ull, .num_endpoints = 1, - .endpoints = &spk_l_endpoint, + .endpoints = &spk_r_endpoint, .name_prefix = "Right" }, { .adr = 0x000127019F837300ull, .num_endpoints = 1, - .endpoints = &spk_r_endpoint, + .endpoints = &spk_l_endpoint, .name_prefix = "Left" } };
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 5/22 ASoC: wm8731: Disable the regulator when probing fails
In reply to Sasha LevinFrom: Zheyu Ma <[email protected]>
[ Upstream commit 92ccbf17eeacf510cf1eed9c252d9332ca24f02d ]
When the driver fails during probing, the driver should disable the regulator, not just handle it in wm8731_hw_init().
The following log reveals it:
[ 17.812483] WARNING: CPU: 1 PID: 364 at drivers/regulator/core.c:2257 _regulator_put+0x3ec/0x4e0 [ 17.815958] RIP: 0010:_regulator_put+0x3ec/0x4e0 [ 17.824467] Call Trace: [ 17.824774] <TASK> [ 17.825040] regulator_bulk_free+0x82/0xe0 [ 17.825514] devres_release_group+0x319/0x3d0 [ 17.825882] i2c_device_probe+0x766/0x940 [ 17.829198] i2c_register_driver+0xb5/0x130
--- sound/soc/codecs/wm8731.c | 19 +++++++++++-------- 1 file changed, 11 insertions(+), 8 deletions(-)
sound/soc/codecs/wm8731.c +11 -8
@@ -602,7 +602,7 @@ static int wm8731_hw_init(struct device *dev, struct wm8731_priv *wm8731) ret =wm8731_reset (wm8731->regmap);if (ret <0 ) {dev_err (dev,"Failed to issue reset: %d\n" , ret); - goto err_regulator_enable; + goto err; }/* Clear POWEROFF, keep everything else disabled */
@@ -619,10 +619,7 @@ static int wm8731_hw_init(struct device *dev, struct wm8731_priv *wm8731)regcache_mark_dirty (wm8731->regmap); -err_regulator_enable: -/* Regulators will be enabled by bias management */ -regulator_bulk_disable (ARRAY_SIZE (wm8731->supplies), wm8731->supplies); - +err:return ret; }
@@ -760,21 +757,27 @@ static int wm8731_i2c_probe(struct i2c_client *i2c, ret =PTR_ERR (wm8731->regmap);dev_err (&i2c->dev,"Failed to allocate register map: %d\n" , ret); -return ret; + goto err_regulator_enable; } ret =wm8731_hw_init (&i2c->dev, wm8731);if (ret !=0 ) -return ret; + goto err_regulator_enable; ret =devm_snd_soc_register_component (&i2c->dev, &soc_component_dev_wm8731, &wm8731_dai,1 );if (ret !=0 ) {dev_err (&i2c->dev,"Failed to register CODEC: %d\n" , ret); -return ret; + goto err_regulator_enable; }return 0 ; + +err_regulator_enable: +/* Regulators will be enabled by bias management */ +regulator_bulk_disable (ARRAY_SIZE (wm8731->supplies), wm8731->supplies); + +return ret; }static int wm8731_i2c_remove (struct i2c_client *client)
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 6/22 ASoC: Intel: sof_es8336: Add a quirk for Huawei Matebook D15
In reply to Sasha LevinFrom: Mauro Carvalho Chehab <[email protected]>
[ Upstream commit c7cb4717f641db68e8117635bfcf62a9c27dc8d3 ]
Based on experimental tests, Huawei Matebook D15 actually uses both gpio0 and gpio1: the first one controls the speaker, while the other one controls the headphone.
Also, the headset is mapped as MIC1, instead of MIC2.
So, add a quirk for it.
--- sound/soc/intel/boards/sof_es8336.c | 9 +++++++++ 1 file changed, 9 insertions(+)
sound/soc/intel/boards/sof_es8336.c +9 -0
@@ -252,6 +252,15 @@ static const struct dmi_system_id sof_es8336_quirk_table[] = { SOF_ES8336_TGL_GPIO_QUIRK | SOF_ES8336_ENABLE_DMIC) }, + { + .callback = sof_es8336_quirk_cb, + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "HUAWEI"), + DMI_MATCH(DMI_BOARD_NAME, "BOHB-WAX9-PCB-B2"), + }, + .driver_data = (void *)(SOF_ES8336_HEADPHONE_GPIO | + SOC_ES8336_HEADSET_MIC1) + }, {} };
Signature · 1 line
2.35.1
Re: [PATCH AUTOSEL 5.17 06/22] ASoC: Intel: sof_es8336: Add a quirk for Huawei Matebook D15
Quoted 7 lines · Sasha Levin
From: Mauro Carvalho Chehab <[email protected]>
[ Upstream commit
c7cb4717f641db68e8117635bfcf62a9c27dc8d3]Based on experimental tests, Huawei Matebook D15 actually uses both gpio0 and gpio1: the first one controls the speaker, while the other one controls the headphone.
Are you sure this doesn't need the rest of the series it came along with?
Sasha Levin
maintainerRe: [PATCH AUTOSEL 5.17 06/22] ASoC: Intel: sof_es8336: Add a quirk for Huawei Matebook D15
Quoted 11 lines · Mark Brown
Quoted 7 lines · Sasha Levin
From: Mauro Carvalho Chehab <[email protected]>
[ Upstream commit
c7cb4717f641db68e8117635bfcf62a9c27dc8d3]Based on experimental tests, Huawei Matebook D15 actually uses both gpio0 and gpio1: the first one controls the speaker, while the other one controls the headphone.
Are you sure this doesn't need the rest of the series it came along with?
I'm not :) Should we queue it too?
Signature · 2 lines
Thanks, Sasha
Re: [PATCH AUTOSEL 5.17 06/22] ASoC: Intel: sof_es8336: Add a quirk for Huawei Matebook D15
Quoted 6 lines · Sasha Levin
Quoted 5 lines · Mark Brown
Quoted 4 lines · Sasha Levin
From: Mauro Carvalho Chehab <[email protected]>
[ Upstream commit
c7cb4717f641db68e8117635bfcf62a9c27dc8d3]
Quoted 8 lines
Quoted 6 lines
Quoted 3 lines
Based on experimental tests, Huawei Matebook D15 actually uses both gpio0 and gpio1: the first one controls the speaker, while the other one controls the headphone.
Are you sure this doesn't need the rest of the series it came along with?
I'm not :) Should we queue it too?
If you add this platform to -stable, you'd need the entire series https://lore.kernel.org/alsa-devel/[email protected]/, and the additions made by Mauro.
My take is that it's not really relevant for -stable, support for this hardware codec is still a works-in-progress and while we'd certainly want to have more distributions use the hardware support it's quite disruptive for -stable maintainers, with the risk of compilation problems and functional issues introduced. it'll make more sense for 5.18+.
Sasha Levin
maintainerPATCH 7/22 Input: cypress-sf - register a callback to disable the regulators
In reply to Sasha LevinFrom: Zheyu Ma <[email protected]>
[ Upstream commit fd0a4b39870d49ff15f6966470185409e261f20f ]
When the driver fails to probe, we will get the following splat:
[ 19.311970] ------------[ cut here ]------------ [ 19.312566] WARNING: CPU: 3 PID: 375 at drivers/regulator/core.c:2257 _regulator_put+0x3ec/0x4e0 [ 19.317591] RIP: 0010:_regulator_put+0x3ec/0x4e0 [ 19.328831] Call Trace: [ 19.329112] <TASK> [ 19.329369] regulator_bulk_free+0x82/0xe0 [ 19.329860] devres_release_group+0x319/0x3d0 [ 19.330357] i2c_device_probe+0x766/0x940
Fix this by adding a callback that will deal with the disabling when the driver fails to probe.
--- drivers/input/keyboard/cypress-sf.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
drivers/input/keyboard/cypress-sf.c +14 -0
@@ -61,6 +61,14 @@ static irqreturn_t cypress_sf_irq_handler(int irq, void *devid)return IRQ_HANDLED ; } +static void cypress_sf_disable_regulators (void *arg) +{ +struct cypress_sf_data *touchkey = arg; + +regulator_bulk_disable (ARRAY_SIZE (touchkey->regulators), + touchkey->regulators); +} +static int cypress_sf_probe (struct i2c_client *client) {struct cypress_sf_data *touchkey;
@@ -121,6 +129,12 @@ static int cypress_sf_probe(struct i2c_client *client)return error; } + error =devm_add_action_or_reset (&client->dev, + cypress_sf_disable_regulators, + touchkey); +if (error) +return error; + touchkey->input_dev =devm_input_allocate_device (&client->dev);if (!touchkey->input_dev) {dev_err (&client->dev,"Failed to allocate input device\n" );
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 8/22 ext4: fix bug_on in start_this_handle during umount filesystem
In reply to Sasha LevinFrom: Ye Bin <[email protected]>
[ Upstream commit b98535d091795a79336f520b0708457aacf55c67 ]
We got issue as follows:
------------[ cut here ]------------ kernel BUG at fs/jbd2/transaction.c:389! invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI CPU: 9 PID: 131 Comm: kworker/9:1 Not tainted 5.17.0-862.14.0.6.x86_64-00001-g23f87daf7d74-dirty #197 Workqueue: events flush_stashed_error_work RIP: 0010:start_this_handle+0x41c/0x1160 RSP: 0018:ffff888106b47c20 EFLAGS: 00010202 RAX: ffffed10251b8400 RBX: ffff888128dc204c RCX: ffffffffb52972ac RDX: 0000000000000200 RSI: 0000000000000004 RDI: ffff888128dc2050 RBP: 0000000000000039 R08: 0000000000000001 R09: ffffed10251b840a R10: ffff888128dc204f R11: ffffed10251b8409 R12: ffff888116d78000 R13: 0000000000000000 R14: dffffc0000000000 R15: ffff888128dc2000 FS: 0000000000000000(0000) GS:ffff88839d680000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000001620068 CR3: 0000000376c0e000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> jbd2__journal_start+0x38a/0x790 jbd2_journal_start+0x19/0x20 flush_stashed_error_work+0x110/0x2b3 process_one_work+0x688/0x1080 worker_thread+0x8b/0xc50 kthread+0x26f/0x310 ret_from_fork+0x22/0x30 </TASK> Modules linked in: ---[ end trace 0000000000000000 ]---
Above issue may happen as follows:
umount read procfs error_workext4_put_super
flush_work(&sbi->s_error_work);
ext4_mb_seq_groups_show ext4_mb_load_buddy_gfp ext4_mb_init_group ext4_mb_init_cache ext4_read_block_bitmap_nowait ext4_validate_block_bitmap ext4_error ext4_handle_error schedule_work(&EXT4_SB(sb)->s_error_work);
ext4_unregister_sysfs(sb);
jbd2_journal_destroy(sbi->s_journal);
journal_kill_thread
journal->j_flags |= JBD2_UNMOUNT;flush_stashed_error_work jbd2_journal_start start_this_handle BUG_ON(journal->j_flags & JBD2_UNMOUNT);
To solve this issue, we call 'ext4_unregister_sysfs() before flushing s_error_work in ext4_put_super().
--- fs/ext4/super.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-)
fs/ext4/super.c +12 -7
@@ -1199,20 +1199,25 @@ static void ext4_put_super(struct super_block *sb)int aborted =0 ;int i, err; -ext4_unregister_li_request (sb); -ext4_quota_off_umount (sb); - -flush_work (&sbi->s_error_work); -destroy_workqueue (sbi->rsv_conversion_wq); -ext4_release_orphan_info (sb); -/* * Unregister sysfs before destroying jbd2 journal. * Since we could still access attr_journal_task attribute via sysfs * path which could have sbi->s_journal->j_task as NULL +* Unregister sysfs before flush sbi->s_error_work. +* Since user may read /proc/fs/ext4/xx/mb_groups during umount, If +* read metadata verify failed then will queue error work. +* flush_stashed_error_work will call start_this_handle may trigger +* BUG_ON. */ ext4_unregister_sysfs (sb); +ext4_unregister_li_request (sb); +ext4_quota_off_umount (sb); + +flush_work (&sbi->s_error_work); +destroy_workqueue (sbi->rsv_conversion_wq); +ext4_release_orphan_info (sb); +if (sbi->s_journal) { aborted =is_journal_aborted (sbi->s_journal); err =jbd2_journal_destroy (sbi->s_journal);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 9/22 arch: xtensa: platforms: Fix deadlock in rs_close()
In reply to Sasha LevinFrom: Duoming Zhou <[email protected]>
[ Upstream commit eb5adc70754d26a260f8b42d39db42da0d0af500 ]
There is a deadlock in rs_close(), which is shown below:
(Thread 1) | (Thread 2)
| rs_open()
rs_close() | mod_timer()
spin_lock_bh() //(1) | (wait a time)
... | rs_poll()
del_timer_sync() | spin_lock() //(2)
(wait timer to stop) | ...We hold timer_lock in position (1) of thread 1 and use del_timer_sync() to wait timer to stop, but timer handler also need timer_lock in position (2) of thread 2. As a result, rs_close() will block forever.
This patch deletes the redundant timer_lock in order to prevent the deadlock. Because there is no race condition between rs_close, rs_open and rs_poll.
Message-Id: <[email protected]>
--- arch/xtensa/platforms/iss/console.c | 8 -------- 1 file changed, 8 deletions(-)
arch/xtensa/platforms/iss/console.c +0 -8
@@ -36,24 +36,19 @@ static void rs_poll(struct timer_list *);static struct tty_driver *serial_driver;static struct tty_port serial_port; static DEFINE_TIMER(serial_timer, rs_poll); -static DEFINE_SPINLOCK (timer_lock);static int rs_open (struct tty_struct *tty,struct file * filp) { -spin_lock_bh (&timer_lock);if (tty->count ==1 )mod_timer (&serial_timer, jiffies +SERIAL_TIMER_VALUE ); -spin_unlock_bh (&timer_lock);return 0 ; }static void rs_close (struct tty_struct *tty,struct file * filp) { -spin_lock_bh (&timer_lock);if (tty->count ==1 )del_timer_sync (&serial_timer); -spin_unlock_bh (&timer_lock); }
@@ -73,8 +68,6 @@ static void rs_poll(struct timer_list *unused)int rd =1 ;unsigned char c; -spin_lock (&timer_lock); -while (simc_poll (0 )) { rd =simc_read (0 , &c,1 );if (rd <=0 )
@@ -87,7 +80,6 @@ static void rs_poll(struct timer_list *unused)tty_flip_buffer_push (port);if (rd)mod_timer (&serial_timer, jiffies +SERIAL_TIMER_VALUE ); -spin_unlock (&timer_lock); }
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 10/22 ksmbd: increment reference count of parent fp
In reply to Sasha LevinFrom: Namjae Jeon <[email protected]>
[ Upstream commit 8510a043d334ecdf83d4604782f288db6bf21d60 ]
Add missing increment reference count of parent fp in ksmbd_lookup_fd_inode().
--- fs/ksmbd/smb2pdu.c | 2 ++ fs/ksmbd/vfs_cache.c | 1 + 2 files changed, 3 insertions(+)
fs/ksmbd/smb2pdu.c +2 -0
@@ -5771,8 +5771,10 @@ static int set_rename_info(struct ksmbd_work *work, struct ksmbd_file *fp,if (parent_fp) {if (parent_fp->daccess &FILE_DELETE_LE ) {pr_err ("parent dir is opened with delete access\n" ); +ksmbd_fd_put (work, parent_fp);return -ESHARE ; } +ksmbd_fd_put (work, parent_fp); } next: return smb2_rename(work, fp, user_ns, rename_info,
fs/ksmbd/vfs_cache.c +1 -0
@@ -497,6 +497,7 @@ struct ksmbd_file *ksmbd_lookup_fd_inode(struct inode *inode)list_for_each_entry (lfp, &ci->m_fp_list, node) {if (inode ==file_inode (lfp->filp)) {atomic_dec (&ci->m_count); + lfp =ksmbd_fp_get (lfp);read_unlock (&ci->m_lock);return lfp; }
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 11/22 ksmbd: set fixed sector size to FS_SECTOR_SIZE_INFORMATION
In reply to Sasha LevinFrom: Namjae Jeon <[email protected]>
[ Upstream commit 02655a70b7cc0f534531ee65fa72692f4d31a944 ]
Currently ksmbd is using ->f_bsize from vfs_statfs() as sector size. If fat/exfat is a local share, ->f_bsize is a cluster size that is too large to be used as a sector size. Sector sizes larger than 4K cause problem occurs when mounting an iso file through windows client.
The error message can be obtained using Mount-DiskImage command,
the error is:
"Mount-DiskImage : The sector size of the physical disk on which the virtual disk resides is not supported."
This patch reports fixed 4KB sector size if ->s_blocksize is bigger than 4KB.
--- fs/ksmbd/smb2pdu.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)
fs/ksmbd/smb2pdu.c +7 -4
@@ -11,6 +11,7 @@#include <linux/statfs.h> #include <linux/ethtool.h> #include <linux/falloc.h> +#include <linux/mount.h> #include "glob.h" #include "smbfsctl.h"
@@ -5005,15 +5006,17 @@ static int smb2_get_info_filesystem(struct ksmbd_work *work,case FS_SECTOR_SIZE_INFORMATION : {struct smb3_fs_ss_info *info; +unsigned int sector_size = + min_t(unsigned int, path.mnt->mnt_sb->s_blocksize, 4096); info = (struct smb3_fs_ss_info *)(rsp->Buffer); - info->LogicalBytesPerSector =cpu_to_le32 (stfs.f_bsize); + info->LogicalBytesPerSector =cpu_to_le32 (sector_size); info->PhysicalBytesPerSectorForAtomicity = -cpu_to_le32 (stfs.f_bsize); - info->PhysicalBytesPerSectorForPerf =cpu_to_le32 (stfs.f_bsize); +cpu_to_le32 (sector_size); + info->PhysicalBytesPerSectorForPerf =cpu_to_le32 (sector_size); info->FSEffPhysicalBytesPerSectorForAtomicity = -cpu_to_le32 (stfs.f_bsize); +cpu_to_le32 (sector_size); info->Flags =cpu_to_le32 (SSINFO_FLAGS_ALIGNED_DEVICE |SSINFO_FLAGS_PARTITION_ALIGNED_ON_DEVICE ); info->ByteOffsetForSectorAlignment =0 ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 12/22 ip6_gre: Avoid updating tunnel->tun_hlen in __gre6_xmit()
In reply to Sasha LevinFrom: Peilin Ye <[email protected]>
[ Upstream commit f40c064e933d7787ca7411b699504d7a2664c1f5 ]
Do not update tunnel->tun_hlen in data plane code. Use a local variable instead, just like "tunnel_hlen" in net/ipv4/ip_gre.c:gre_fb_xmit().
--- net/ipv6/ip6_gre.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
net/ipv6/ip6_gre.c +3 -2
@@ -743,6 +743,7 @@ static netdev_tx_t __gre6_xmit(struct sk_buff *skb,struct ip_tunnel_info *tun_info;const struct ip_tunnel_key *key;__be16 flags; +int tun_hlen; tun_info =skb_tunnel_info_txcheck (skb); if (IS_ERR(tun_info) ||
@@ -760,9 +761,9 @@ static netdev_tx_t __gre6_xmit(struct sk_buff *skb, dsfield = key->tos; flags = key->tun_flags & (TUNNEL_CSUM |TUNNEL_KEY |TUNNEL_SEQ ); - tunnel->tun_hlen =gre_calc_hlen (flags); + tun_hlen =gre_calc_hlen (flags); - gre_build_header(skb, tunnel->tun_hlen, + gre_build_header(skb, tun_hlen, flags, protocol, tunnel_id_to_key32(tun_info->key.tun_id), (flags & TUNNEL_SEQ) ? htonl(tunnel->o_seqno++)
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 13/22 erofs: fix use-after-free of on-stack io[]
In reply to Sasha LevinFrom: Hongyu Jin <[email protected]>
[ Upstream commit 60b30050116c0351b90154044345c1b53ae1f323 ]
The root cause is the race as follows:
Thread #1 Thread #2(irq ctx)
z_erofs_runqueue()
struct z_erofs_decompressqueue io_A[];
submit bio A
z_erofs_decompress_kickoff(,,1)
z_erofs_decompressqueue_endio(bio A)
z_erofs_decompress_kickoff(,,-1)
spin_lock_irqsave()
atomic_add_return()
io_wait_event() -> pending_bios is already 0
[end of function]
wake_up_locked(io_A[]) // crashReferenced backtrace in kernel 5.4:
[ 10.129422] Unable to handle kernel paging request at virtual address eb0454a4 [ 10.364157] CPU: 0 PID: 709 Comm: getprop Tainted: G WC O 5.4.147-ab09225 #1 [ 11.556325] [<c01b33b8>] (__wake_up_common) from [<c01b3300>] (__wake_up_locked+0x40/0x48) [ 11.565487] [<c01b3300>] (__wake_up_locked) from [<c044c8d0>] (z_erofs_vle_unzip_kickoff+0x6c/0xc0) [ 11.575438] [<c044c8d0>] (z_erofs_vle_unzip_kickoff) from [<c044c854>] (z_erofs_vle_read_endio+0x16c/0x17c) [ 11.586082] [<c044c854>] (z_erofs_vle_read_endio) from [<c06a80e8>] (clone_endio+0xb4/0x1d0) [ 11.595428] [<c06a80e8>] (clone_endio) from [<c04a1280>] (blk_update_request+0x150/0x4dc) [ 11.604516] [<c04a1280>] (blk_update_request) from [<c06dea28>] (mmc_blk_cqe_complete_rq+0x144/0x15c) [ 11.614640] [<c06dea28>] (mmc_blk_cqe_complete_rq) from [<c04a5d90>] (blk_done_softirq+0xb0/0xcc) [ 11.624419] [<c04a5d90>] (blk_done_softirq) from [<c010242c>] (__do_softirq+0x184/0x56c) [ 11.633419] [<c010242c>] (__do_softirq) from [<c01051e8>] (irq_exit+0xd4/0x138) [ 11.641640] [<c01051e8>] (irq_exit) from [<c010c314>] (__handle_domain_irq+0x94/0xd0) [ 11.650381] [<c010c314>] (__handle_domain_irq) from [<c04fde70>] (gic_handle_irq+0x50/0xd4) [ 11.659641] [<c04fde70>] (gic_handle_irq) from [<c0101b70>] (__irq_svc+0x70/0xb0)
--- fs/erofs/zdata.c | 12 ++++-------- fs/erofs/zdata.h | 2 +- 2 files changed, 5 insertions(+), 9 deletions(-)
fs/erofs/zdata.c +4 -8
@@ -1073,12 +1073,9 @@ static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io,/* wake up the caller thread for sync decompression */ if (sync) { -unsigned long flags; - -spin_lock_irqsave (&io->u.wait.lock, flags);if (!atomic_add_return (bios, &io->pending_bios)) -wake_up_locked (&io->u.wait); -spin_unlock_irqrestore (&io->u.wait.lock, flags); +complete (&io->u.done); +return ; }
@@ -1224,7 +1221,7 @@ jobqueue_init(struct super_block *sb, }else { fg_out: q = fgq; -init_waitqueue_head (&fgq->u.wait); +init_completion (&fgq->u.done);atomic_set (&fgq->pending_bios,0 ); } q->sb = sb;
@@ -1428,8 +1425,7 @@ static void z_erofs_runqueue(struct super_block *sb,return ;/* wait until all bios are completed */ -io_wait_event (io[JQ_SUBMIT ].u.wait, - !atomic_read (&io[JQ_SUBMIT ].pending_bios)); +wait_for_completion_io (&io[JQ_SUBMIT ].u.done);/* handle synchronous decompress queue in the caller context */ z_erofs_decompress_queue (&io[JQ_SUBMIT ], pagepool);
fs/erofs/zdata.h +1 -1
@@ -97,7 +97,7 @@ struct z_erofs_decompressqueue {z_erofs_next_pcluster_t head;union { -wait_queue_head_t wait; +struct completion done;struct work_struct work; } u; };
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 14/22 bonding: do not discard lowest hash bit for non layer3+4 hashing
In reply to Sasha LevinFrom: suresh kumar <[email protected]>
[ Upstream commit 49aefd131739df552f83c566d0665744c30b1d70 ]
Commit b5f862180d70 was introduced to discard lowest hash bit for layer3+4 hashing but it also removes last bit from non layer3+4 hashing
Below script shows layer2+3 hashing will result in same slave to be used with above commit.
$ cat hash.py #/usr/bin/python3.6
h_dests=[0xa0, 0xa1]
h_source=0xe3
hproto=0x8
saddr=0x1e7aa8c0
daddr=0x17aa8c0
for h_dest in h_dests:
hash = (h_dest ^ h_source ^ hproto ^ saddr ^ daddr)
hash ^= hash >> 16
hash ^= hash >> 8
print(hash)print("with last bit removed")
for h_dest in h_dests:
hash = (h_dest ^ h_source ^ hproto ^ saddr ^ daddr) hash ^= hash >> 16
hash ^= hash >> 8
hash = hash >> 1
print(hash)
Output:
$ python3.6 hash.py 522133332 522133333 <-------------- will result in both slaves being used
with last bit removed
261066666
261066666 <-------------- only single slave used
--- drivers/net/bonding/bond_main.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/net/bonding/bond_main.c +9 -4
@@ -3819,14 +3819,19 @@ static bool bond_flow_dissect(struct bonding *bond, struct sk_buff *skb, const vreturn true; } -static u32 bond_ip_hash (u32 hash,struct flow_keys *flow) +static u32 bond_ip_hash (u32 hash,struct flow_keys *flow,int xmit_policy) { hash ^= (__force u32 )flow_get_u32_dst (flow) ^ (__force u32 )flow_get_u32_src (flow); hash ^= (hash >>16 ); hash ^= (hash >>8 ); +/* discard lowest hash bit to deal with the common even ports pattern */ -return hash >>1 ; +if (xmit_policy ==BOND_XMIT_POLICY_LAYER34 || + xmit_policy ==BOND_XMIT_POLICY_ENCAP34 ) +return hash >>1 ; + +return hash; }/* Generate hash based on xmit policy. If @skb is given it is used to linearize
@@ -3856,7 +3861,7 @@ static u32 __bond_xmit_hash(struct bonding *bond, struct sk_buff *skb, const voimemcpy (&hash, &flow.ports.ports,sizeof (hash)); } -return bond_ip_hash (hash, &flow); +return bond_ip_hash (hash, &flow, bond->params.xmit_policy); }/**
@@ -5051,7 +5056,7 @@ static u32 bond_sk_hash_l34(struct sock *sk)/* L4 */ memcpy (&hash, &flow.ports.ports,sizeof (hash));/* L3 */ -return bond_ip_hash (hash, &flow); +return bond_ip_hash (hash, &flow,BOND_XMIT_POLICY_LAYER34 ); } static struct net_device *__bond_sk_get_lower_dev(struct bonding *bond,
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 15/22 x86: __memcpy_flushcache: fix wrong alignment if size > 2^32
In reply to Sasha LevinFrom: Mikulas Patocka <[email protected]>
[ Upstream commit a6823e4e360fe975bd3da4ab156df7c74c8b07f3 ]
The first "if" condition in __memcpy_flushcache is supposed to align the "dest" variable to 8 bytes and copy data up to this alignment. However, this condition may misbehave if "size" is greater than 4GiB.
The statement min_t(unsigned, size, ALIGN(dest, 8) - dest); casts both arguments to unsigned int and selects the smaller one. However, the cast truncates high bits in "size" and it results in misbehavior.
For example:
suppose that size == 0x100000001, dest == 0x200000002 min_t(unsigned, size, ALIGN(dest, 8) - dest) == min_t(0x1, 0xe) == 0x1; ... dest += 0x1;
so we copy just one byte "and" dest remains unaligned.
This patch fixes the bug by replacing unsigned with size_t.
--- arch/x86/lib/usercopy_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/lib/usercopy_64.c +1 -1
@@ -119,7 +119,7 @@ void __memcpy_flushcache(void *_dst, const void *_src, size_t size)/* cache copy and flush to align dest */ if (!IS_ALIGNED (dest,8 )) { -unsigned len =min_t (unsigned, size,ALIGN (dest,8 ) - dest); +size_t len =min_t (size_t, size,ALIGN (dest,8 ) - dest);memcpy ((void *) dest, (void *) source, len);clean_cache_range ((void *) dest, len);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 16/22 cifs: destage any unwritten data to the server before calling copychunk_write
In reply to Sasha LevinFrom: Ronnie Sahlberg <[email protected]>
[ Upstream commit f5d0f921ea362636e4a2efb7c38d1ead373a8700 ]
because the copychunk_write might cover a region of the file that has not yet been sent to the server and thus fail.
A simple way to reproduce this is:
truncate -s 0 /mnt/testfile; strace -f -o x -ttT xfs_io -i -f -c 'pwrite 0k 128k' -c 'fcollapse 16k 24k' /mnt/testfile
the issue is that the 'pwrite 0k 128k' becomes rearranged on the wire with the 'fcollapse 16k 24k' due to write-back caching.
fcollapse is implemented in cifs.ko as a SMB2 IOCTL(COPYCHUNK_WRITE) call and it will fail serverside since the file is still 0b in size serverside until the writes have been destaged.
To avoid this we must ensure that we destage any unwritten data to the server before calling COPYCHUNK_WRITE.
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1997373
--- fs/cifs/smb2ops.c | 8 ++++++++ 1 file changed, 8 insertions(+)
fs/cifs/smb2ops.c +8 -0
@@ -1861,9 +1861,17 @@ smb2_copychunk_range(const unsigned int xid,int chunks_copied =0 ;bool chunk_sizes_updated = false;ssize_t bytes_written, total_bytes_written =0 ; +struct inode *inode; pcchunk =kmalloc (sizeof (struct copychunk_ioctl ),GFP_KERNEL ); +/* +* We need to flush all unwritten data before we can send the +* copychunk ioctl to the server. +*/ + inode =d_inode (trgtfile->dentry); +filemap_write_and_wait (inode->i_mapping); +if (pcchunk ==NULL )return -ENOMEM ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 17/22 drivers: net: hippi: Fix deadlock in rr_close()
In reply to Sasha LevinFrom: Duoming Zhou <[email protected]>
[ Upstream commit bc6de2878429e85c1f1afaa566f7b5abb2243eef ]
There is a deadlock in rr_close(), which is shown below:
(Thread 1) | (Thread 2)
| rr_open()
rr_close() | add_timer()
spin_lock_irqsave() //(1) | (wait a time)
... | rr_timer()
del_timer_sync() | spin_lock_irqsave() //(2)
(wait timer to stop) | ...We hold rrpriv->lock in position (1) of thread 1 and use del_timer_sync() to wait timer to stop, but timer handler also need rrpriv->lock in position (2) of thread 2. As a result, rr_close() will block forever.
This patch extracts del_timer_sync() from the protection of spin_lock_irqsave(), which could let timer handler to obtain the needed lock.
--- drivers/net/hippi/rrunner.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/hippi/rrunner.c +2 -0
@@ -1355,7 +1355,9 @@ static int rr_close(struct net_device *dev) rrpriv->fw_running =0 ; +spin_unlock_irqrestore (&rrpriv->lock, flags);del_timer_sync (&rrpriv->timer); +spin_lock_irqsave (&rrpriv->lock, flags);writel (0 , ®s->TxPi);writel (0 , ®s->IpRxPi);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 18/22 powerpc/perf: Fix 32bit compile
In reply to Sasha LevinFrom: Alexey Kardashevskiy <[email protected]>
[ Upstream commit bb82c574691daf8f7fa9a160264d15c5804cb769 ]
The "read_bhrb" global symbol is only called under CONFIG_PPC64 of arch/powerpc/perf/core-book3s.c but it is compiled for both 32 and 64 bit anyway (and LLVM fails to link this on 32bit).
This fixes it by moving bhrb.o to obj64 targets.
--- arch/powerpc/perf/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/perf/Makefile +2 -2
@@ -3,11 +3,11 @@ obj-y += callchain.o callchain_$(BITS).o perf_regs.o obj-$(CONFIG_COMPAT) += callchain_32.o -obj-$(CONFIG_PPC_PERF_CTRS) += core-book3s.o bhrb.o +obj-$(CONFIG_PPC_PERF_CTRS) += core-book3s.o obj64-$(CONFIG_PPC_PERF_CTRS) += ppc970-pmu.o power5-pmu.o \ power5+-pmu.o power6-pmu.o power7-pmu.o \ isa207-common.o power8-pmu.o power9-pmu.o \ - generic-compat-pmu.o power10-pmu.o + generic-compat-pmu.o power10-pmu.o bhrb.o obj32-$(CONFIG_PPC_PERF_CTRS) += mpc7450-pmu.o obj-$(CONFIG_PPC_POWERNV) += imc-pmu.o
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 19/22 ALSA: hda: intel-dsp-config: Add RaptorLake PCI IDs
In reply to Sasha LevinFrom: Gongjun Song <[email protected]>
[ Upstream commit b07908ab26ceab51165c13714277c19252e62594 ]
Add RaptorLake-P PCI IDs
--- sound/hda/intel-dsp-config.c | 9 +++++++++ 1 file changed, 9 insertions(+)
sound/hda/intel-dsp-config.c +9 -0
@@ -410,6 +410,15 @@ static const struct config_entry config_table[] = { .flags = FLAG_SOF | FLAG_SOF_ONLY_IF_DMIC_OR_SOUNDWIRE, .device = 0x54c8, }, + /* RaptorLake-P */ + { + .flags = FLAG_SOF | FLAG_SOF_ONLY_IF_DMIC_OR_SOUNDWIRE, + .device = 0x51ca, + }, + { + .flags = FLAG_SOF | FLAG_SOF_ONLY_IF_DMIC_OR_SOUNDWIRE, + .device = 0x51cb, + }, #endif };
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 20/22 ata: pata_marvell: Check the 'bmdma_addr' beforing reading
In reply to Sasha LevinFrom: Zheyu Ma <[email protected]>
[ Upstream commit aafa9f958342db36c17ac2a7f1b841032c96feb4 ]
Before detecting the cable type on the dma bar, the driver should check whether the 'bmdma_addr' is zero, which means the adapter does not support DMA, otherwise we will get the following error:
[ 5.146634] Bad IO access at port 0x1 (return inb(port)) [ 5.147206] WARNING: CPU: 2 PID: 303 at lib/iomap.c:44 ioread8+0x4a/0x60 [ 5.150856] RIP: 0010:ioread8+0x4a/0x60 [ 5.160238] Call Trace: [ 5.160470] <TASK> [ 5.160674] marvell_cable_detect+0x6e/0xc0 [pata_marvell] [ 5.161728] ata_eh_recover+0x3520/0x6cc0 [ 5.168075] ata_do_eh+0x49/0x3c0
--- drivers/ata/pata_marvell.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/ata/pata_marvell.c +2 -0
@@ -77,6 +77,8 @@ static int marvell_cable_detect(struct ata_port *ap)switch (ap->port_no) {case 0 : +if (!ap->ioaddr.bmdma_addr) +return ATA_CBL_PATA_UNK ;if (ioread8 (ap->ioaddr.bmdma_addr +1 ) &1 )return ATA_CBL_PATA40 ;return ATA_CBL_PATA80 ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 21/22 selftest/vm: verify mmap addr in mremap_test
In reply to Sasha LevinFrom: Sidhartha Kumar <[email protected]>
[ Upstream commit 9c85a9bae267f6b5e5e374d0d023bbbe9db096d3 ]
Avoid calling mmap with requested addresses that are less than the system's mmap_min_addr. When run as root, mmap returns EACCES when trying to map addresses < mmap_min_addr. This is not one of the error codes for the condition to retry the mmap in the test.
Rather than arbitrarily retrying on EACCES, don't attempt an mmap until addr > vm.mmap_min_addr.
Add a munmap call after an alignment check as the mappings are retained after the retry and can reach the vm.max_map_count sysctl.
--- tools/testing/selftests/vm/mremap_test.c | 41 +++++++++++++++++++++++- 1 file changed, 40 insertions(+), 1 deletion(-)
tools/testing/selftests/vm/mremap_test.c +40 -1
@@ -6,6 +6,7 @@#include <errno.h> #include <stdlib.h> +#include <stdio.h> #include <string.h> #include <sys/mman.h> #include <time.h>
@@ -63,6 +64,35 @@ enum { .expect_failure = should_fail \ } +/* Returns mmap_min_addr sysctl tunable from procfs */ +static unsigned long long get_mmap_min_addr (void ) +{ +FILE *fp; +int n_matched; +static unsigned long long addr; + +if (addr) +return addr; + + fp =fopen ("/proc/sys/vm/mmap_min_addr" ,"r" ); +if (fp ==NULL ) { +ksft_print_msg ("Failed to open /proc/sys/vm/mmap_min_addr: %s\n" , +strerror (errno)); +exit (KSFT_SKIP ); + } + + n_matched =fscanf (fp,"%llu" , &addr); +if (n_matched !=1 ) { +ksft_print_msg ("Failed to read /proc/sys/vm/mmap_min_addr: %s\n" , +strerror (errno)); +fclose (fp); +exit (KSFT_SKIP ); + } + +fclose (fp); +return addr; +} +/* * Returns the start address of the mapping on success, else returns * NULL on failure.
@@ -71,8 +101,15 @@ static void *get_source_mapping(struct config c) {unsigned long long addr =0ULL ;void *src_addr =NULL ; +unsigned long long mmap_min_addr; + + mmap_min_addr =get_mmap_min_addr (); + retry: addr += c.src_alignment; +if (addr < mmap_min_addr) + goto retry; + src_addr =mmap ((void *) addr, c.region_size,PROT_READ |PROT_WRITE ,MAP_FIXED_NOREPLACE |MAP_ANONYMOUS |MAP_SHARED ,-1 ,0 );
@@ -90,8 +127,10 @@ static void *get_source_mapping(struct config c)* alignment in the tests. */ if (((unsigned long long ) src_addr & (c.src_alignment -1 )) || - !((unsigned long long ) src_addr & c.src_alignment)) + !((unsigned long long ) src_addr & c.src_alignment)) { +munmap (src_addr, c.region_size); goto retry; + }if (!src_addr) goto error;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 22/22 selftest/vm: verify remap destination address in mremap_test
In reply to Sasha LevinFrom: Sidhartha Kumar <[email protected]>
[ Upstream commit 18d609daa546c919fd36b62a7b510c18de4b4af8 ]
Because mremap does not have a MAP_FIXED_NOREPLACE flag, it can destroy existing mappings. This causes a segfault when regions such as text are remapped and the permissions are changed.
Verify the requested mremap destination address does not overlap any existing mappings by using mmap's MAP_FIXED_NOREPLACE flag. Keep incrementing the destination address until a valid mapping is found or fail the current test once the max address is reached.
--- tools/testing/selftests/vm/mremap_test.c | 42 ++++++++++++++++++++++-- 1 file changed, 39 insertions(+), 3 deletions(-)
tools/testing/selftests/vm/mremap_test.c +39 -3
@@ -10,6 +10,7 @@#include <string.h> #include <sys/mman.h> #include <time.h> +#include <stdbool.h> #include "../kselftest.h"
@@ -64,6 +65,30 @@ enum { .expect_failure = should_fail \ } +/* +* Returns false if the requested remap region overlaps with an +* existing mapping (e.g text, stack) else returns true. +*/ +static bool is_remap_region_valid (void *addr,unsigned long long size) +{ +void *remap_addr =NULL ; +bool ret = true; + +/* Use MAP_FIXED_NOREPLACE flag to ensure region is not mapped */ + remap_addr =mmap (addr, size,PROT_READ |PROT_WRITE , +MAP_FIXED_NOREPLACE |MAP_ANONYMOUS |MAP_SHARED , +-1 ,0 ); + +if (remap_addr ==MAP_FAILED ) { +if (errno ==EEXIST ) + ret = false; + }else { +munmap (remap_addr, size); + } + +return ret; +} +/* Returns mmap_min_addr sysctl tunable from procfs */ static unsigned long long get_mmap_min_addr (void ) {
@@ -111,8 +136,8 @@ static void *get_source_mapping(struct config c) goto retry; src_addr =mmap ((void *) addr, c.region_size,PROT_READ |PROT_WRITE , -MAP_FIXED_NOREPLACE |MAP_ANONYMOUS |MAP_SHARED , --1 ,0 ); +MAP_FIXED_NOREPLACE |MAP_ANONYMOUS |MAP_SHARED , +-1 ,0 );if (src_addr ==MAP_FAILED ) {if (errno ==EPERM || errno ==EEXIST ) goto retry;
@@ -179,9 +204,20 @@ static long long remap_region(struct config c, unsigned int threshold_mb,if (!((unsigned long long ) addr & c.dest_alignment)) addr = (void *) ((unsigned long long ) addr | c.dest_alignment); +/* Don't destroy existing mappings unless expected to overlap */ +while (!is_remap_region_valid (addr, c.region_size) && !c.overlapping) { +/* Check for unsigned overflow */ +if (addr + c.dest_alignment < addr) { +ksft_print_msg ("Couldn't find a valid region to remap to\n" ); + ret =-1 ; + goto out; + } + addr += c.dest_alignment; + } +clock_gettime (CLOCK_MONOTONIC , &t_start); dest_addr =mremap (src_addr, c.region_size, c.region_size, -MREMAP_MAYMOVE |MREMAP_FIXED , (char *) addr); +MREMAP_MAYMOVE |MREMAP_FIXED , (char *) addr);clock_gettime (CLOCK_MONOTONIC , &t_end);if (dest_addr ==MAP_FAILED ) {
Signature · 1 line
2.35.1
Hi,
I don't think that this patch needs backporting.
It does not fix anything and could introduce some regression if b7898396f4bb is not also already backported.
It could avoid some (unlikely?) allocation failure, but as this case is already handled ("if (!dpcm)"), it shouldn't be an issue if it happened.
Just for my understanding, why has it auto-selected for backport?
I thought that a Fixes tag and/or a real reported issue was need for this to happen.
CJ
Le 26/04/2022 à 21:01, Sasha Levin a écrit :
Quoted 35 lines
From: Christophe JAILLET <[email protected]>
[ Upstream commit
fb6d679fee95d272c0a94912c4e534146823ee89]At the
kzalloc()call indpcm_be_connect(), there is no spin lock involved. It's merely protected bycard->pcm_mutex, instead. The spinlock is applied at the later call withsnd_soc_pcm_stream_lock_irq()only for the list manipulations. (See it's*_irq(), not*_irqsave(); that means the context being sleepable at that point.) So, we can useGFP_KERNELsafely there.This patch revert commit
d8a9c6e1f676("ASoC: soc-pcm: useGFP_ATOMICfor dpcm structure") which is no longer needed since commitb7898396f4bb("ASoC: soc-pcm: Fix and cleanup DPCM locking").Signed-off-by Christophe JAILLET <[email protected]>Link https://lore.kernel.org/r/e740f1930843060e025e3c0f17ec1393cfdafb26.1648757961.git.christophe.jaillet@wanadoo.fr open hereSigned-off-by Mark Brown <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- sound/soc/soc-pcm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)sound/soc/soc-pcm.c +1
-1@@ -1214,7 +1214,7 @@ static int dpcm_be_connect(struct snd_soc_pcm_runtime *fe, be_substream->pcm->nonatomic =1 ; } - dpcm =kzalloc (sizeof (struct snd_soc_dpcm ),GFP_ATOMIC ); + dpcm =kzalloc (sizeof (struct snd_soc_dpcm ),GFP_KERNEL );if (!dpcm)return -ENOMEM ;