Commit graph

1,296,507 commits

Author SHA1 Message Date
Justin Stitt
855ce06f91 scsi: wd33c93: Replace deprecated strncpy() with strscpy()
@p1 is assigned to @setup_buffer and then we manually assign a NUL-byte at
the first index. This renders the following strlen() call useless.
Moreover, we don't need to reassign p1 to setup_buffer for any reason --
neither do we need to manually set a NUL-byte at the end. strscpy()
resolves all this code making it easier to read.

Even considering the path where @str is falsey, the manual NUL-byte
assignment is useless as setup_buffer is declared with static storage
duration in the top-level scope which should NUL-initialize the whole
buffer.

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Link: https://lore.kernel.org/r/20240305-strncpy-drivers-scsi-mpi3mr-mpi3mr_fw-c-v3-7-5b78a13ff984@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:37:43 -04:00
Justin Stitt
8fd4c9c8e1 scsi: smartpqi: Replace deprecated strncpy() with strscpy()
buffer->driver_version is sized 32:
|	struct bmic_host_wellness_driver_version {
|	...
|		char	driver_version[32];
... the source string "Linux " + DRIVER_VERISON is sized at 16. There's
really no bug in the existing code since the buffers are sized
appropriately with great care taken to manually NUL-terminate the
destination buffer. Nonetheless, let's make the swap over to strscpy()
for robustness' (and readability's) sake.

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Link: https://lore.kernel.org/r/20240305-strncpy-drivers-scsi-mpi3mr-mpi3mr_fw-c-v3-6-5b78a13ff984@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:37:43 -04:00
Justin Stitt
1b60c86dd9 scsi: devinfo: Replace strncpy() and manual pad
Depending on the state of @compatible, we are going to do different things
with our @to buffer.

When @compatible is true we want a NUL-term'd and NUL-padded destination
buffer. Conversely, if @compatible is false we just want a space-padded
destination buffer (no NUL-term required).

As per:
/**
 * scsi_dev_info_list_add_keyed - add one dev_info list entry.
 * @compatible: if true, null terminate short strings.  Otherwise space pad.
...

Note that we can't easily use strtomem_pad() here as the size of the @to
buffer is unknown to the compiler due to indirection layers.

Now, the intent of the code is more clear (I probably didn't even need
to add a comment -- that's how clear it is).

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Link: https://lore.kernel.org/r/20240305-strncpy-drivers-scsi-mpi3mr-mpi3mr_fw-c-v3-5-5b78a13ff984@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:37:43 -04:00
Justin Stitt
4f94864d21 scsi: qla4xxx: Replace deprecated strncpy() with strscpy()
Replace 3 instances of strncpy in ql4_mbx.c

No bugs exist in the current implementation as some care was taken to
ensure the write length was decreased by one to leave some space for a
NUL-byte. However, instead of using strncpy(dest, src, LEN-1) we can opt
for strscpy(dest, src, sizeof(dest)) which will result in NUL-termination
as well. It should be noted that the entire chap_table is zero-allocated so
the NUL-padding provided by strncpy is not needed.

While here, I noticed that MIN_CHAP_SECRET_LEN was not used anywhere.
Since strscpy gives us the number of bytes copied into the destination
buffer (or an -E2BIG) we can check both for an error during copying and
also for a non-length compliant secret. Add a new jump label so we can
properly clean up our chap_table should we have to abort due to bad secret.

The third instance in this file involves some more peculiar handling of
strings:
|	uint32_t mbox_cmd[MBOX_REG_COUNT];
|	...
|	memset(&mbox_cmd, 0, sizeof(mbox_cmd));
|	...
|	mbox_cmd[0] = MBOX_CMD_SET_PARAM;
|	if (param == SET_DRVR_VERSION) {
|		mbox_cmd[1] = SET_DRVR_VERSION;
|		strncpy((char *)&mbox_cmd[2], QLA4XXX_DRIVER_VERSION,
|			MAX_DRVR_VER_LEN - 1);

mbox_cmd has a size of 8:
|	#define MBOX_REG_COUNT 8
... and its type width is 4 bytes. Hence, we have 32 bytes to work with
here. The first 4 bytes are used as a flag for the MBOX_CMD_SET_PARAM.
The next 4 bytes are used for SET_DRVR_VERSION. We now have 32-8=24
bytes remaining -- which thankfully is what MAX_DRVR_VER_LEN is equal to
|	#define MAX_DRVR_VER_LEN                    24

... and the thing we're copying into this pseudo-string buffer is
|	#define QLA4XXX_DRIVER_VERSION        "5.04.00-k6"

... which is great because its less than 24 bytes (therefore we aren't
truncating the source).

All to say, there's no bug in the existing implementation (yay!) but we
can clean the code up a bit by using strscpy().

In ql4_os.c, there aren't any strncpy() uses to replace but there are
some existing strscpy() calls that could be made more idiomatic. Where
possible, use strscpy(dest, src, sizeof(dest)). Note that
chap_rec->password has a size of ISCSI_CHAP_AUTH_SECRET_MAX_LEN
|	#define ISCSI_CHAP_AUTH_SECRET_MAX_LEN	256
... while the current strscpy usage uses QL4_CHAP_MAX_SECRET_LEN
|	#define QL4_CHAP_MAX_SECRET_LEN 100
... however since chap_table->secret was set and bounded properly in its
string assignment its probably safe here to switch over to sizeof().

|	struct iscsi_chap_rec {
	...
|		char username[ISCSI_CHAP_AUTH_NAME_MAX_LEN];
|		uint8_t password[ISCSI_CHAP_AUTH_SECRET_MAX_LEN];
	...
|	};

|	strscpy(chap_rec->password, chap_table->secret,
|		QL4_CHAP_MAX_SECRET_LEN);

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Link: https://lore.kernel.org/r/20240305-strncpy-drivers-scsi-mpi3mr-mpi3mr_fw-c-v3-4-5b78a13ff984@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:37:43 -04:00
Justin Stitt
2303149d58 scsi: qedf: Replace deprecated strncpy() with strscpy()
We expect slowpath_params.name to be NUL-terminated based on its future
usage with other string APIs:

|	static int qed_slowpath_start(struct qed_dev *cdev,
|				      struct qed_slowpath_params *params)
...
|	strscpy(drv_version.name, params->name,
|		MCP_DRV_VER_STR_SIZE - 4);

Moreover, NUL-padding is not necessary as the only use for this slowpath
name parameter is to copy into the drv_version.name field.

Also, let's prefer using strscpy(src, dest, sizeof(src)) in two instances
(one of which is outside of the scsi system but it is trivial and related
to this patch).

We can see the drv_version.name size here:
|	struct qed_mcp_drv_version {
|		u32	version;
|		u8	name[MCP_DRV_VER_STR_SIZE - 4];
|	};

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Link: https://lore.kernel.org/r/20240305-strncpy-drivers-scsi-mpi3mr-mpi3mr_fw-c-v3-3-5b78a13ff984@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:37:43 -04:00
Justin Stitt
b7e9712a02 scsi: mpt3sas: Replace deprecated strncpy() with strscpy()
The replacement in mpt3sas_base.c is a trivial one because desc is already
zero-initialized meaning there is no functional change here.

For mpt3sas_transport.c, we know edev is zero-initialized as well while
manufacture_reply comes from dma_alloc_coherent(). No functional change
here either.

For all cases, use the more idiomatic strscpy() usage of: strscpy(dest,
src, sizeof(dest))

Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Link: https://lore.kernel.org/r/20240305-strncpy-drivers-scsi-mpi3mr-mpi3mr_fw-c-v3-2-5b78a13ff984@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:37:43 -04:00
Justin Stitt
8c8e2422bd scsi: mpi3mr: Replace deprecated strncpy() with assignments
Really, there's no bug with the current code. Let's just ditch strncpy()
all together.

We can just copy the const strings instead of reserving room on the stack.

Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1]
Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [2]
Link: https://github.com/KSPP/linux/issues/90
Cc: linux-hardening@vger.kernel.org
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Justin Stitt <justinstitt@google.com>
Link: https://lore.kernel.org/r/20240305-strncpy-drivers-scsi-mpi3mr-mpi3mr_fw-c-v3-1-5b78a13ff984@google.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:37:43 -04:00
Muhammad Usama Anjum
16cc2ba71b scsi: lpfc: Correct size for cmdwqe/rspwqe for memset()
The cmdwqe and rspwqe are of type lpfc_wqe128. They should be memset() with
the same type.

Fixes: 61910d6a52 ("scsi: lpfc: SLI path split: Refactor CT paths")
Signed-off-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Link: https://lore.kernel.org/r/20240304091119.847060-1-usama.anjum@collabora.com
Reviewed-by: Justin Tee <justin.tee@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:17:32 -04:00
Muhammad Usama Anjum
28d4199118 scsi: lpfc: Correct size for wqe for memset()
The wqe is of type lpfc_wqe128. It should be memset with the same type.

Fixes: 6c621a2229 ("scsi: lpfc: Separate NVMET RQ buffer posting from IO resources SGL/iocbq/context")
Signed-off-by: Muhammad Usama Anjum <usama.anjum@collabora.com>
Link: https://lore.kernel.org/r/20240304090649.833953-1-usama.anjum@collabora.com
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Justin Tee <justintee8345@gmail.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:16:58 -04:00
Ricardo B. Marliere
0822853d65 scsi: st: Make st_sysfs_class constant
Since commit 43a7206b09 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the st_sysfs_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Link: https://lore.kernel.org/r/20240302-class_cleanup-scsi-v1-5-b9096b990e27@marliere.net
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:15:49 -04:00
Ricardo B. Marliere
ac9f3ac5b8 scsi: ch: Make ch_sysfs_class constant
Since commit 43a7206b09 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the ch_sysfs_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Link: https://lore.kernel.org/r/20240302-class_cleanup-scsi-v1-4-b9096b990e27@marliere.net
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:15:48 -04:00
Ricardo B. Marliere
a08f0eb029 scsi: cxlflash: Make cxlflash_class constant
Since commit 43a7206b09 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the cxlflash_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Link: https://lore.kernel.org/r/20240302-class_cleanup-scsi-v1-3-b9096b990e27@marliere.net
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:15:48 -04:00
Ricardo B. Marliere
ee8dda6a7e scsi: pmcraid: Make pmcraid_class constant
Since commit 43a7206b09 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the pmcraid_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Link: https://lore.kernel.org/r/20240302-class_cleanup-scsi-v1-2-b9096b990e27@marliere.net
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:15:48 -04:00
Ricardo B. Marliere
f1fb41765d scsi: sg: Make sg_sysfs_class constant
Since commit 43a7206b09 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the sg_sysfs_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.

Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Link: https://lore.kernel.org/r/20240302-class_cleanup-scsi-v1-1-b9096b990e27@marliere.net
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:15:48 -04:00
Peter Wang
db06ae7ce9 scsi: ufs: core: Add config_scsi_dev vops comment
Add config_scsi_dev vops comment.

Signed-off-by: Peter Wang <peter.wang@mediatek.com>
Link: https://lore.kernel.org/r/20240301034610.24928-1-peter.wang@mediatek.com
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:10:24 -04:00
Dmitry Baryshkov
81e2c1a0f8 scsi: ufs: qcom: Provide default cycles_in_1us value
The MSM8996 DT doesn't provide frequency limits for the core_clk_unipro
clock, which results in miscalculation of the cycles_in_1us value.  Provide
the backwards-compatible default to support existing MSM8996 DT files.

Fixes: b4e13e1ae9 ("scsi: ufs: qcom: Add multiple frequency support for MAX_CORE_CLK_1US_CYCLES")
Cc: Nitin Rawat <quic_nitirawa@quicinc.com>
Cc: stable@vger.kernel.org # 6.7.x
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20240218-msm8996-fix-ufs-v3-1-40aab49899a3@linaro.org
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:09:29 -04:00
Xingui Yang
8e68a458bc scsi: libsas: Fix disk not being scanned in after being removed
As of commit d8649fc1c5 ("scsi: libsas: Do discovery on empty PHY to
update PHY info"), do discovery will send a new SMP_DISCOVER and update
phy->phy_change_count. We found that if the disk is reconnected and phy
change_count changes at this time, the disk scanning process will not be
triggered.

Therefore, call sas_set_ex_phy() to update the PHY info with the results of
the last query. And because the previous phy info will be used when calling
sas_unregister_devs_sas_addr(), sas_unregister_devs_sas_addr() should be
called before sas_set_ex_phy().

Fixes: d8649fc1c5 ("scsi: libsas: Do discovery on empty PHY to update PHY info")
Signed-off-by: Xingui Yang <yangxingui@huawei.com>
Link: https://lore.kernel.org/r/20240307141413.48049-3-yangxingui@huawei.com
Reviewed-by: John Garry <john.g.garry@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:02:54 -04:00
Xingui Yang
a57345279f scsi: libsas: Add a helper sas_get_sas_addr_and_dev_type()
Add a helper to get attached_sas_addr and device type from disc_resp.

Suggested-by: John Garry <john.g.garry@oracle.com>
Signed-off-by: Xingui Yang <yangxingui@huawei.com>
Link: https://lore.kernel.org/r/20240307141413.48049-2-yangxingui@huawei.com
Reviewed-by: John Garry <john.g.garry@oracle.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:02:53 -04:00
Colin Ian King
99cfb212ef scsi: target: iscsi: Remove unused variable xfer_len
The variable 'xfer_len' is being initialized and incremented but it is
never actually referenced in any other way. The variable is redundant and
can be removed.

Cleans up clang scan build warning:
drivers/target/iscsi/iscsi_target_erl1.c:586:45: warning: variable
'xfer_len' set but not used [-Wunused-but-set-variable]

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Link: https://lore.kernel.org/r/20240307104553.1980860-1-colin.i.king@gmail.com
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 18:01:34 -04:00
Duje Mihanović
2d77f70bb7 Input: imagis - add touch key support
IST3032C (and possibly some other models) has touch keys. Add support
for them to the imagis driver.

Signed-off-by: Duje Mihanović <duje.mihanovic@skole.hr>
Link: https://lore.kernel.org/r/20240306-b4-imagis-keys-v3-3-2c429afa8420@skole.hr
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
2024-03-10 14:38:04 -07:00
Duje Mihanović
44b6cee041 dt-bindings: input: imagis: Document touch keys
IST3032C (and possibly some other models) has touch keys. Document this.

Signed-off-by: Duje Mihanović <duje.mihanovic@skole.hr>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20240306-b4-imagis-keys-v3-2-2c429afa8420@skole.hr
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
2024-03-10 14:38:04 -07:00
Duje Mihanović
c0ca3dbd03 Input: imagis - use FIELD_GET where applicable
Instead of manually extracting certain bits from registers with binary
ANDs and shifts, the FIELD_GET macro can be used. With this in mind, the
*_SHIFT macros can be dropped.

Signed-off-by: Duje Mihanović <duje.mihanovic@skole.hr>
Link: https://lore.kernel.org/r/20240306-b4-imagis-keys-v3-1-2c429afa8420@skole.hr
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
2024-03-10 14:38:04 -07:00
Ricardo B. Marliere
a4735d40a5 Input: make input_class constant
Since commit 43a7206b09 ("driver core: class: make class_register() take
a const *"), the driver core allows for struct class to be in read-only
memory, so move the input_class structure to be declared at build time
placing it into read-only memory, instead of having to be dynamically
allocated at boot time.

Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net>
Link: https://lore.kernel.org/r/20240305-class_cleanup-input-v1-1-0c3d950c25db@marliere.net
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
2024-03-10 14:38:04 -07:00
Daniel Golle
b8a77b9a5f mtd: ubi: fix NVMEM over UBI volumes on 32-bit systems
A compiler warning related to sizeof(int) != 8 when calling do_div()
is triggered when building on 32-bit platforms.
Address this by using integer types having a well-defined size.

Fixes: 3ce485803d ("mtd: ubi: provide NVMEM layer over UBI volumes")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Zhihao Cheng <chengzhihao1@huawei.com>
Tested-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Richard Weinberger <richard@nod.at>
2024-03-10 22:14:28 +01:00
Rohit Ner
767712f91d scsi: ufs: core: Fix MCQ MAC configuration
As per JEDEC Standard No. 223E Section 5.9.2, the max # active commands
value programmed by the host sw in MCQConfig.MAC should be one less than
the actual value.

Signed-off-by: Rohit Ner <rohitner@google.com>
Link: https://lore.kernel.org/r/20240220095637.2900067-1-rohitner@google.com
Reviewed-by: Peter Wang <peter.wang@mediatek.com>
Reviewed-by: Can Guo <quic_cang@quicinc.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
2024-03-10 17:10:15 -04:00
Linus Torvalds
e8f897f4af Linux 6.8 v6.8 2024-03-10 13:38:09 -07:00
Armin Wolf
8debe3c129 hwmon: (dell-smm) Add XPS 9315 to fan control whitelist
A user reported that on this machine, disabling BIOS fan control
is necessary in order to change the fan speed.

Signed-off-by: Armin Wolf <W_Armin@gmx.de>
Acked-by: Pali Rohár <pali@kernel.org>
Link: https://lore.kernel.org/r/20240309212025.13758-1-W_Armin@gmx.de
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
2024-03-10 12:58:18 -07:00
Kent Overstreet
737cd174d1 bcachefs: bch2_lookup() gives better error message on inode not found
When a dirent points to a missing inode, we really should print out the
dirent.

This requires quite a bit of refactoring, but there's some other
benefits: we now do the entire looup (dirent and inode) in a single
btree transaction, and copy to the VFS inode with btree locks still
held, like the create path.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
a91bc5e505 bcachefs: bch2_inode_insert()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
eab0af905b mm: introduce PF_MEMALLOC_NORECLAIM, PF_MEMALLOC_NOWARN
Introduce PF_MEMALLOC_* equivalents of some GFP_ flags:

PF_MEMALLOC_NORECLAIM	-> GFP_NOWAIT
PF_MEMALLOC_NOWARN	-> __GFP_NOWARN

Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Darrick J. Wong <djwong@kernel.org>
Cc: linux-mm@kvack.org
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
3f6d5e6a46 mm: introduce memalloc_flags_{save,restore}
Our proliferation of memalloc_*_{save,restore} APIs is getting a bit
silly, this adds a generic version and converts the existing
save/restore functions to wrappers.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Darrick J. Wong <djwong@kernel.org>
Cc: linux-mm@kvack.org
Acked-by: Vlastimil Babka <vbabka@suse.cz>
2024-03-10 15:34:09 -04:00
Kent Overstreet
3d4998c202 bcachefs: factor out check_inode_backpointer()
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
11def1888f bcachefs: Factor out check_subvol_dirent()
Going to be adding more code here for checking subvol structure.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
ce3e9283de bcachefs: Kill some -EINVALs
Repurposing standard error codes in bcachefs code is banned in new code,
and we need to get rid of the remaining ones - private error codes give
us much better error messages.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
82fdc1dc98 bcachefs: bump max_active on btree_interior_update_worker
WQ_UNBOUND with max_active 1 means ordered workqueue, but we don't
actually need or want ordered semantics - and probably want a higher
concurrency limit anyways.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
69c8e6ce02 bcachefs: move fsck_write_inode() to inode.c
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
29223b5a55 bcachefs: Initialize super_block->s_uuid
Need to fix this oversight for the new FS_IOC_(GET|SET)UUID ioctls.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
f8f8fb443b bcachefs: Switch to uuid_to_fsid()
switch the statfs code from something horrible and open coded to the
more standard uuid_to_fsid()

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:09 -04:00
Kent Overstreet
7f76b08aca bcachefs: Subvolumes may now be renamed
Files within a subvolume cannot be renamed into another subvolume, but
subvolumes themselves were intended to be.

This implements subvolume renaming - we need to ensure that there's only
a single dirent that points to a subvolume key (not multiple versions in
different snapshots), and we need to ensure that dirent.d_parent_subol
and inode.bi_parent_subvol are updated.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
5f43b0134e bcachefs: btree node prefetching in check_topology
btree_and_journal_iter is old code that we want to get rid of, but we're
not ready to yet.

lack of btree node prefetching is, it turns out, a real performance
issue for fsck on spinning rust, so - add it.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
fc634d8e46 bcachefs: btree_and_journal_iter.trans
we now always have a btree_trans when using a btree_and_journal_iter;
prep work for adding prefetching to btree_and_journal_iter

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
916abefd43 bcachefs: better journal pipelining
Recently a severe performance regression was discovered, which bisected
to

  a6548c8b5e bcachefs: Avoid flushing the journal in the discard path

It turns out the old behaviour, which issued excessive journal flushes,
worked around a performance issue where queueing delays would cause the
journal to not be able to write quickly enough and stall.

The journal flushes masked the issue because they periodically flushed
the device write cache, reducing write latency for non flushes.

This patch reworks the journalling code to allow more than one
(non-flush) write to be in flight at a time. With this patch, doing 4k
random writes and an iodepth of 128, we are now able to hit 560k iops to
a Samsung 970 EVO Plus - previously, we were stuck in the ~200k range.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
38789c2508 bcachefs: closure per journal buf
Prep work for having multiple journal writes in flight.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
5165400275 bcachefs: bio per journal buf
Prep work for having multiple journal writes in flight.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
52f7d75e7d bcachefs: jset_entry_datetime
This gives us a way to record the date and time every journal entry was
written - useful for debugging.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
3d3d23b341 bcachefs: improve journal entry read fsck error messages
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
a555bcf4fa bcachefs: convert journal replay ptrs to darray
Eliminates some error paths - no longer have a hardcoded
BCH_REPLICAS_MAX limit.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
5b6271b509 bcachefs: Cleanup bch2_dirent_lookup_trans()
Drop an unnecessary bch2_subvolume_get_snapshot() call, and drop the __
from the name - this is a normal interface.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
23f2522315 bcachefs: bch2_hash_set_snapshot() -> bch2_hash_set_in_snapshot()
Minor renaming for clarity, bit of refactoring.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00
Kent Overstreet
6b83aee8a4 bcachefs: Workqueues should be WQ_HIGHPRI
Most bcachefs workqueues are used for completions, and should be
WQ_HIGHPRI - this helps reduce queuing delays, we want to complete
quickly once we can no longer signal backpressure by blocking.

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2024-03-10 15:34:08 -04:00