In the scenario of entering hibernation with udisk in the system, if the
udisk was gone or resume fail in the thaw phase of hibernation. Its state
will be set to NOTATTACHED. At this point, usb_hub_wq was already freezed
and can't not handle disconnect event. Next, in the poweroff phase of
hibernation, SYNCHRONIZE_CACHE SCSI command will be sent to this udisk
when poweroff this scsi device, which will cause uas_submit_urbs to be
called to submit URB for sense/data/cmd pipe. However, these URBs will
submit fail as device was set to NOTATTACHED state. Then, uas_submit_urbs
will return a value SCSI_MLQUEUE_DEVICE_BUSY to the caller. That will lead
the SCSI layer go into an ugly loop and system fail to go into hibernation.
On the other hand, when we specially check for -ENODEV in function
uas_queuecommand_lck, returning DID_ERROR to SCSI layer will cause device
poweroff fail and system shutdown instead of entering hibernation.
To fix this issue, let uas_submit_urbs to return original generic error
when submitting URB failed. At the same time, we need to translate -ENODEV
to DID_NOT_CONNECT for the SCSI layer.
Suggested-by: Oliver Neukum <oneukum(a)suse.com>
Cc: stable(a)vger.kernel.org
Signed-off-by: Weitao Wang <WeitaoWang-oc(a)zhaoxin.com>
---
v3->v4
- remove unused variable declaration in function uas_submit_urbs.
drivers/usb/storage/uas.c | 28 +++++++++++++---------------
1 file changed, 13 insertions(+), 15 deletions(-)
diff --git a/drivers/usb/storage/uas.c b/drivers/usb/storage/uas.c
index 9707f53cfda9..5930cfc03111 100644
--- a/drivers/usb/storage/uas.c
+++ b/drivers/usb/storage/uas.c
@@ -533,7 +533,7 @@ static struct urb *uas_alloc_cmd_urb(struct uas_dev_info *devinfo, gfp_t gfp,
* daft to me.
*/
-static struct urb *uas_submit_sense_urb(struct scsi_cmnd *cmnd, gfp_t gfp)
+static int uas_submit_sense_urb(struct scsi_cmnd *cmnd, gfp_t gfp)
{
struct uas_dev_info *devinfo = cmnd->device->hostdata;
struct urb *urb;
@@ -541,30 +541,28 @@ static struct urb *uas_submit_sense_urb(struct scsi_cmnd *cmnd, gfp_t gfp)
urb = uas_alloc_sense_urb(devinfo, gfp, cmnd);
if (!urb)
- return NULL;
+ return -ENOMEM;
usb_anchor_urb(urb, &devinfo->sense_urbs);
err = usb_submit_urb(urb, gfp);
if (err) {
usb_unanchor_urb(urb);
uas_log_cmd_state(cmnd, "sense submit err", err);
usb_free_urb(urb);
- return NULL;
}
- return urb;
+ return err;
}
static int uas_submit_urbs(struct scsi_cmnd *cmnd,
struct uas_dev_info *devinfo)
{
struct uas_cmd_info *cmdinfo = scsi_cmd_priv(cmnd);
- struct urb *urb;
int err;
lockdep_assert_held(&devinfo->lock);
if (cmdinfo->state & SUBMIT_STATUS_URB) {
- urb = uas_submit_sense_urb(cmnd, GFP_ATOMIC);
- if (!urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ err = uas_submit_sense_urb(cmnd, GFP_ATOMIC);
+ if (err)
+ return err;
cmdinfo->state &= ~SUBMIT_STATUS_URB;
}
@@ -572,7 +570,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
cmdinfo->data_in_urb = uas_alloc_data_urb(devinfo, GFP_ATOMIC,
cmnd, DMA_FROM_DEVICE);
if (!cmdinfo->data_in_urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return -ENOMEM;
cmdinfo->state &= ~ALLOC_DATA_IN_URB;
}
@@ -582,7 +580,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (err) {
usb_unanchor_urb(cmdinfo->data_in_urb);
uas_log_cmd_state(cmnd, "data in submit err", err);
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return err;
}
cmdinfo->state &= ~SUBMIT_DATA_IN_URB;
cmdinfo->state |= DATA_IN_URB_INFLIGHT;
@@ -592,7 +590,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
cmdinfo->data_out_urb = uas_alloc_data_urb(devinfo, GFP_ATOMIC,
cmnd, DMA_TO_DEVICE);
if (!cmdinfo->data_out_urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return -ENOMEM;
cmdinfo->state &= ~ALLOC_DATA_OUT_URB;
}
@@ -602,7 +600,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (err) {
usb_unanchor_urb(cmdinfo->data_out_urb);
uas_log_cmd_state(cmnd, "data out submit err", err);
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return err;
}
cmdinfo->state &= ~SUBMIT_DATA_OUT_URB;
cmdinfo->state |= DATA_OUT_URB_INFLIGHT;
@@ -611,7 +609,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (cmdinfo->state & ALLOC_CMD_URB) {
cmdinfo->cmd_urb = uas_alloc_cmd_urb(devinfo, GFP_ATOMIC, cmnd);
if (!cmdinfo->cmd_urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return -ENOMEM;
cmdinfo->state &= ~ALLOC_CMD_URB;
}
@@ -621,7 +619,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (err) {
usb_unanchor_urb(cmdinfo->cmd_urb);
uas_log_cmd_state(cmnd, "cmd submit err", err);
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return err;
}
cmdinfo->cmd_urb = NULL;
cmdinfo->state &= ~SUBMIT_CMD_URB;
@@ -698,7 +696,7 @@ static int uas_queuecommand_lck(struct scsi_cmnd *cmnd)
* of queueing, no matter how fatal the error
*/
if (err == -ENODEV) {
- set_host_byte(cmnd, DID_ERROR);
+ set_host_byte(cmnd, DID_NO_CONNECT);
scsi_done(cmnd);
goto zombie;
}
--
2.32.0
In the scenario of entering hibernation with udisk in the system, if the
udisk was gone or resume fail in the thaw phase of hibernation. Its state
will be set to NOTATTACHED. At this point, usb_hub_wq was already freezed
and can't not handle disconnect event. Next, in the poweroff phase of
hibernation, SYNCHRONIZE_CACHE SCSI command will be sent to this udisk
when poweroff this scsi device, which will cause uas_submit_urbs to be
called to submit URB for sense/data/cmd pipe. However, these URBs will
submit fail as device was set to NOTATTACHED state. Then, uas_submit_urbs
will return a value SCSI_MLQUEUE_DEVICE_BUSY to the caller. That will lead
the SCSI layer go into an ugly loop and system fail to go into hibernation.
On the other hand, when we specially check for -ENODEV in function
uas_queuecommand_lck, returning DID_ERROR to SCSI layer will cause device
poweroff fail and system shutdown instead of entering hibernation.
To fix this issue, let uas_submit_urbs to return original generic error
when submitting URB failed. At the same time, we need to translate -ENODEV
to DID_NOT_CONNECT for the SCSI layer.
Suggested-by: Oliver Neukum <oneukum(a)suse.com>
Cc: stable(a)vger.kernel.org
Signed-off-by: Weitao Wang <WeitaoWang-oc(a)zhaoxin.com>
---
v2->v3
- Modify the description of this patch.
- An error is returned directly when submitting URB fails.
drivers/usb/storage/uas.c | 27 +++++++++++++--------------
1 file changed, 13 insertions(+), 14 deletions(-)
diff --git a/drivers/usb/storage/uas.c b/drivers/usb/storage/uas.c
index 9707f53cfda9..689396777b6f 100644
--- a/drivers/usb/storage/uas.c
+++ b/drivers/usb/storage/uas.c
@@ -533,7 +533,7 @@ static struct urb *uas_alloc_cmd_urb(struct uas_dev_info *devinfo, gfp_t gfp,
* daft to me.
*/
-static struct urb *uas_submit_sense_urb(struct scsi_cmnd *cmnd, gfp_t gfp)
+static int uas_submit_sense_urb(struct scsi_cmnd *cmnd, gfp_t gfp)
{
struct uas_dev_info *devinfo = cmnd->device->hostdata;
struct urb *urb;
@@ -541,16 +541,15 @@ static struct urb *uas_submit_sense_urb(struct scsi_cmnd *cmnd, gfp_t gfp)
urb = uas_alloc_sense_urb(devinfo, gfp, cmnd);
if (!urb)
- return NULL;
+ return -ENOMEM;
usb_anchor_urb(urb, &devinfo->sense_urbs);
err = usb_submit_urb(urb, gfp);
if (err) {
usb_unanchor_urb(urb);
uas_log_cmd_state(cmnd, "sense submit err", err);
usb_free_urb(urb);
- return NULL;
}
- return urb;
+ return err;
}
static int uas_submit_urbs(struct scsi_cmnd *cmnd,
@@ -562,9 +561,9 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
lockdep_assert_held(&devinfo->lock);
if (cmdinfo->state & SUBMIT_STATUS_URB) {
- urb = uas_submit_sense_urb(cmnd, GFP_ATOMIC);
- if (!urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ err = uas_submit_sense_urb(cmnd, GFP_ATOMIC);
+ if (err)
+ return err;
cmdinfo->state &= ~SUBMIT_STATUS_URB;
}
@@ -572,7 +571,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
cmdinfo->data_in_urb = uas_alloc_data_urb(devinfo, GFP_ATOMIC,
cmnd, DMA_FROM_DEVICE);
if (!cmdinfo->data_in_urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return -ENOMEM;
cmdinfo->state &= ~ALLOC_DATA_IN_URB;
}
@@ -582,7 +581,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (err) {
usb_unanchor_urb(cmdinfo->data_in_urb);
uas_log_cmd_state(cmnd, "data in submit err", err);
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return err;
}
cmdinfo->state &= ~SUBMIT_DATA_IN_URB;
cmdinfo->state |= DATA_IN_URB_INFLIGHT;
@@ -592,7 +591,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
cmdinfo->data_out_urb = uas_alloc_data_urb(devinfo, GFP_ATOMIC,
cmnd, DMA_TO_DEVICE);
if (!cmdinfo->data_out_urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return -ENOMEM;
cmdinfo->state &= ~ALLOC_DATA_OUT_URB;
}
@@ -602,7 +601,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (err) {
usb_unanchor_urb(cmdinfo->data_out_urb);
uas_log_cmd_state(cmnd, "data out submit err", err);
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return err;
}
cmdinfo->state &= ~SUBMIT_DATA_OUT_URB;
cmdinfo->state |= DATA_OUT_URB_INFLIGHT;
@@ -611,7 +610,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (cmdinfo->state & ALLOC_CMD_URB) {
cmdinfo->cmd_urb = uas_alloc_cmd_urb(devinfo, GFP_ATOMIC, cmnd);
if (!cmdinfo->cmd_urb)
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return -ENOMEM;
cmdinfo->state &= ~ALLOC_CMD_URB;
}
@@ -621,7 +620,7 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
if (err) {
usb_unanchor_urb(cmdinfo->cmd_urb);
uas_log_cmd_state(cmnd, "cmd submit err", err);
- return SCSI_MLQUEUE_DEVICE_BUSY;
+ return err;
}
cmdinfo->cmd_urb = NULL;
cmdinfo->state &= ~SUBMIT_CMD_URB;
@@ -698,7 +697,7 @@ static int uas_queuecommand_lck(struct scsi_cmnd *cmnd)
* of queueing, no matter how fatal the error
*/
if (err == -ENODEV) {
- set_host_byte(cmnd, DID_ERROR);
+ set_host_byte(cmnd, DID_NO_CONNECT);
scsi_done(cmnd);
goto zombie;
}
--
2.32.0
From: Zi Yan <ziy(a)nvidia.com>
The tail pages in a THP can have swap entry information stored in their
private field. When migrating to a new page, all tail pages of the new
page need to update ->private to avoid future data corruption.
This fix is stable-only, since after commit 07e09c483cbe ("mm/huge_memory:
work on folio->swap instead of page->private when splitting folio"),
subpages of a swapcached THP no longer requires the maintenance.
Adding THPs to the swapcache was introduced in commit
38d8b4e6bdc87 ("mm, THP, swap: delay splitting THP during swap out"),
where each subpage of a THP added to the swapcache had its own swapcache
entry and required the ->private field to point to the correct swapcache
entry. Later, when THP migration functionality was implemented in commit
616b8371539a6 ("mm: thp: enable thp migration in generic path"),
it initially did not handle the subpages of swapcached THPs, failing to
update their ->private fields or replace the subpage pointers in the
swapcache. Subsequently, commit e71769ae5260 ("mm: enable thp migration
for shmem thp") addressed the swapcache update aspect. This patch fixes
the update of subpage ->private fields.
Closes: https://lore.kernel.org/linux-mm/1707814102-22682-1-git-send-email-quic_cha…
Fixes: 616b8371539a ("mm: thp: enable thp migration in generic path")
Signed-off-by: Zi Yan <ziy(a)nvidia.com>
Acked-by: David Hildenbrand <david(a)redhat.com>
---
mm/migrate.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/mm/migrate.c b/mm/migrate.c
index c7d5566623ad..c37af50f312d 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -424,8 +424,12 @@ int migrate_page_move_mapping(struct address_space *mapping,
if (PageSwapBacked(page)) {
__SetPageSwapBacked(newpage);
if (PageSwapCache(page)) {
+ int i;
+
SetPageSwapCache(newpage);
- set_page_private(newpage, page_private(page));
+ for (i = 0; i < (1 << compound_order(page)); i++)
+ set_page_private(newpage + i,
+ page_private(page + i));
}
} else {
VM_BUG_ON_PAGE(PageSwapCache(page), page);
--
2.43.0
From: Zi Yan <ziy(a)nvidia.com>
The tail pages in a THP can have swap entry information stored in their
private field. When migrating to a new page, all tail pages of the new
page need to update ->private to avoid future data corruption.
This fix is stable-only, since after commit 07e09c483cbe ("mm/huge_memory:
work on folio->swap instead of page->private when splitting folio"),
subpages of a swapcached THP no longer requires the maintenance.
Adding THPs to the swapcache was introduced in commit
38d8b4e6bdc87 ("mm, THP, swap: delay splitting THP during swap out"),
where each subpage of a THP added to the swapcache had its own swapcache
entry and required the ->private field to point to the correct swapcache
entry. Later, when THP migration functionality was implemented in commit
616b8371539a6 ("mm: thp: enable thp migration in generic path"),
it initially did not handle the subpages of swapcached THPs, failing to
update their ->private fields or replace the subpage pointers in the
swapcache. Subsequently, commit e71769ae5260 ("mm: enable thp migration
for shmem thp") addressed the swapcache update aspect. This patch fixes
the update of subpage ->private fields.
Closes: https://lore.kernel.org/linux-mm/1707814102-22682-1-git-send-email-quic_cha…
Fixes: 616b8371539a ("mm: thp: enable thp migration in generic path")
Signed-off-by: Zi Yan <ziy(a)nvidia.com>
Acked-by: David Hildenbrand <david(a)redhat.com>
---
mm/migrate.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/mm/migrate.c b/mm/migrate.c
index fcb7eb6a6eca..c0a8f3c9e256 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -447,8 +447,12 @@ int migrate_page_move_mapping(struct address_space *mapping,
if (PageSwapBacked(page)) {
__SetPageSwapBacked(newpage);
if (PageSwapCache(page)) {
+ int i;
+
SetPageSwapCache(newpage);
- set_page_private(newpage, page_private(page));
+ for (i = 0; i < (1 << compound_order(page)); i++)
+ set_page_private(newpage + i,
+ page_private(page + i));
}
} else {
VM_BUG_ON_PAGE(PageSwapCache(page), page);
--
2.43.0
From: Zi Yan <ziy(a)nvidia.com>
The tail pages in a THP can have swap entry information stored in their
private field. When migrating to a new page, all tail pages of the new
page need to update ->private to avoid future data corruption.
This fix is stable-only, since after commit 07e09c483cbe ("mm/huge_memory:
work on folio->swap instead of page->private when splitting folio"),
subpages of a swapcached THP no longer requires the maintenance.
Adding THPs to the swapcache was introduced in commit
38d8b4e6bdc87 ("mm, THP, swap: delay splitting THP during swap out"),
where each subpage of a THP added to the swapcache had its own swapcache
entry and required the ->private field to point to the correct swapcache
entry. Later, when THP migration functionality was implemented in commit
616b8371539a6 ("mm: thp: enable thp migration in generic path"),
it initially did not handle the subpages of swapcached THPs, failing to
update their ->private fields or replace the subpage pointers in the
swapcache. Subsequently, commit e71769ae5260 ("mm: enable thp migration
for shmem thp") addressed the swapcache update aspect. This patch fixes
the update of subpage ->private fields.
Closes: https://lore.kernel.org/linux-mm/1707814102-22682-1-git-send-email-quic_cha…
Fixes: 616b8371539a ("mm: thp: enable thp migration in generic path")
Signed-off-by: Zi Yan <ziy(a)nvidia.com>
Acked-by: David Hildenbrand <david(a)redhat.com>
---
mm/migrate.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/mm/migrate.c b/mm/migrate.c
index 034b0662fd3b..9cfd53eaeb4e 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -441,8 +441,12 @@ int migrate_page_move_mapping(struct address_space *mapping,
if (PageSwapBacked(page)) {
__SetPageSwapBacked(newpage);
if (PageSwapCache(page)) {
+ int i;
+
SetPageSwapCache(newpage);
- set_page_private(newpage, page_private(page));
+ for (i = 0; i < (1 << compound_order(page)); i++)
+ set_page_private(newpage + i,
+ page_private(page + i));
}
} else {
VM_BUG_ON_PAGE(PageSwapCache(page), page);
--
2.43.0
Hello everyone,
#regzbot introduced v6.7.5..v6.7.6
I'm experiencing an issue where kexec does a full firmware reboot
instead of kexec reboot.
Issue first submitted at OpenSuse bugzilla [0].
OS details as follows:
Distributor ID: openSUSE
Description: openSUSE Tumbleweed-Slowroll
Release: 20240213
Issue has been reproduced by building kernel from source.
kexec works as expected in kernel v6.7.5.
kexec does full firmware reboot in kernel v6.7.6.
I followed the docs here [1] to perform git bisect and find the culprit,
hope it's alright as I'm quite out of my depth here.
Git bisect logs:
git bisect start
# status: waiting for both good and bad commits
# bad: [b631f5b445dc3379f67ff63a2e4c58f22d4975dc] Linux 6.7.6
git bisect bad b631f5b445dc3379f67ff63a2e4c58f22d4975dc
# status: waiting for good commit(s), bad commit known
# good: [004dcea13dc10acaf1486d9939be4c793834c13c] Linux 6.7.5
git bisect good 004dcea13dc10acaf1486d9939be4c793834c13c
Let me know if there's anything else I can do to help troubleshoot the
issue.
[0]: https://bugzilla.suse.com/show_bug.cgi?id=1220541
[1]: https://docs.kernel.org/admin-guide/bug-bisect.html
Kind regards,
Pavin Joseph.