From: Hou Tao <[email protected]>
When neither no_read_workqueue nor no_write_workqueue are enabled,
tasklet_trylock() in crypt_dec_pending() may still return false due to
an uninitialized state, and dm-crypt will do io completion in io_queue
instead of current context unnecessarily.
Fix it by initializing io->tasklet in crypt_io_init().
Fixes: 8e14f610159d ("dm crypt: do not call bio_endio() from the dm-crypt tasklet")
Signed-off-by: Hou Tao <[email protected]>
---
drivers/md/dm-crypt.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 3aeeb8f2802f..caee6ce3b79f 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -238,6 +238,7 @@ static void crypt_endio(struct bio *clone);
static void kcryptd_queue_crypt(struct dm_crypt_io *io);
static struct scatterlist *crypt_get_sg_data(struct crypt_config *cc,
struct scatterlist *sg);
+static void kcryptd_crypt_tasklet(unsigned long work);
static bool crypt_integrity_aead(struct crypt_config *cc);
@@ -1725,6 +1726,7 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
io->sector = sector;
io->error = 0;
io->ctx.r.req = NULL;
+ tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
io->integrity_metadata = NULL;
io->integrity_metadata_from_pool = false;
atomic_set(&io->io_pending, 0);
@@ -2226,7 +2228,6 @@ static void kcryptd_queue_crypt(struct dm_crypt_io *io)
* it is being executed with irqs disabled.
*/
if (in_hardirq() || irqs_disabled()) {
- tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
tasklet_schedule(&io->tasklet);
return;
}
--
2.29.2
On Mon, Mar 06 2023 at 8:49P -0500,
Hou Tao <[email protected]> wrote:
> From: Hou Tao <[email protected]>
>
> When neither no_read_workqueue nor no_write_workqueue are enabled,
> tasklet_trylock() in crypt_dec_pending() may still return false due to
> an uninitialized state, and dm-crypt will do io completion in io_queue
> instead of current context unnecessarily.
Have you actually experienced this?
> Fix it by initializing io->tasklet in crypt_io_init().
Really would rather avoid always calling tasklet_init(). But I can
optimize it away with a later patch.
Mike
Hi,
On 3/7/2023 3:31 AM, Mike Snitzer wrote:
> On Mon, Mar 06 2023 at 8:49P -0500,
> Hou Tao <[email protected]> wrote:
>
>> From: Hou Tao <[email protected]>
>>
>> When neither no_read_workqueue nor no_write_workqueue are enabled,
>> tasklet_trylock() in crypt_dec_pending() may still return false due to
>> an uninitialized state, and dm-crypt will do io completion in io_queue
>> instead of current context unnecessarily.
> Have you actually experienced this?
Yes. I had written a bpftrace script to check the completion context of
blkdev_bio_end_io_simple() when doing direct io read on dm-crypt device. The
expected context should be unbound workers of crypt_queue, but sometimes the
context is the bound worker of io_queue.
>
>> Fix it by initializing io->tasklet in crypt_io_init().
> Really would rather avoid always calling tasklet_init(). But I can
> optimize it away with a later patch.
My first though was "io->tasklet.state = 0", but it may be fragile because it
operated on the internal status of tasklet, so I switch to tasklet_init().
>
> Mike
On Mon, Mar 06 2023 at 9:12P -0500,
Hou Tao <[email protected]> wrote:
> Hi,
>
> On 3/7/2023 3:31 AM, Mike Snitzer wrote:
> > On Mon, Mar 06 2023 at 8:49P -0500,
> > Hou Tao <[email protected]> wrote:
> >
> >> From: Hou Tao <[email protected]>
> >>
> >> When neither no_read_workqueue nor no_write_workqueue are enabled,
> >> tasklet_trylock() in crypt_dec_pending() may still return false due to
> >> an uninitialized state, and dm-crypt will do io completion in io_queue
> >> instead of current context unnecessarily.
> > Have you actually experienced this?
>
> Yes. I had written a bpftrace script to check the completion context of
> blkdev_bio_end_io_simple() when doing direct io read on dm-crypt device. The
> expected context should be unbound workers of crypt_queue, but sometimes the
> context is the bound worker of io_queue.
OK, thanks for clarifying. Curious to know the circumstance (I
thought per-bio-data is zero'd -- but it may be I'm mistaken).
I won't be marking this commit for stable@ but if others feel
differently please let me know and I'll do so. (We can always propose
it to stable@, after the fact, even if the commit header doesn't Cc
stable@)
> >> Fix it by initializing io->tasklet in crypt_io_init().
> > Really would rather avoid always calling tasklet_init(). But I can
> > optimize it away with a later patch.
>
> My first though was "io->tasklet.state = 0", but it may be fragile because it
> operated on the internal status of tasklet, so I switch to tasklet_init().
Yes, I looked into it and came up with the same hack.. and I too felt
it was too fragile due to open-coding direct access to the tasklet's
members.
I have a patch I just staged that staged that uses jump_labels to
optimize this code. If you might review/test/verify it works well for
you that'd be appreciated:
https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=ae75a25bd83f7c541240449d2fff3a44433e506b
It builds on your patch, which I added a comment to:
https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=d9fe0a98a2e0a1cf585e8a6555afb33be968bd13
From: Mike Snitzer <[email protected]>
Date: Mon, 6 Mar 2023 15:58:33 -0500
Subject: [PATCH] dm crypt: conditionally enable code needed for tasklet usecases
Use jump_label to limit the need for branching, and tasklet_init(),
unless either of the optional "no_read_workqueue" and/or
"no_write_workqueue" features are used.
Signed-off-by: Mike Snitzer <[email protected]>
---
drivers/md/dm-crypt.c | 35 +++++++++++++++++++++++++++--------
1 file changed, 27 insertions(+), 8 deletions(-)
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 641457e72603..2d0309ca07f5 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -40,6 +40,7 @@
#include <keys/user-type.h>
#include <keys/encrypted-type.h>
#include <keys/trusted-type.h>
+#include <linux/jump_label.h>
#include <linux/device-mapper.h>
@@ -85,6 +86,8 @@ struct dm_crypt_io {
struct rb_node rb_node;
} CRYPTO_MINALIGN_ATTR;
+static DEFINE_STATIC_KEY_FALSE(use_tasklet_enabled);
+
struct dm_crypt_request {
struct convert_context *ctx;
struct scatterlist sg_in[4];
@@ -1730,12 +1733,15 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
io->sector = sector;
io->error = 0;
io->ctx.r.req = NULL;
- /*
- * tasklet_init() here to ensure crypt_dec_pending()'s
- * tasklet_trylock() doesn't incorrectly return false
- * even when tasklet isn't in use.
- */
- tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
+ if (static_branch_unlikely(&use_tasklet_enabled)) {
+ /*
+ * tasklet_init() here to ensure crypt_dec_pending()'s
+ * tasklet_trylock() doesn't incorrectly return false
+ * even when tasklet isn't in use.
+ */
+ tasklet_init(&io->tasklet, kcryptd_crypt_tasklet,
+ (unsigned long)&io->work);
+ }
io->integrity_metadata = NULL;
io->integrity_metadata_from_pool = false;
atomic_set(&io->io_pending, 0);
@@ -1775,6 +1781,10 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
kfree(io->integrity_metadata);
base_bio->bi_status = error;
+ if (!static_branch_unlikely(&use_tasklet_enabled)) {
+ bio_endio(base_bio);
+ return;
+ }
/*
* If we are running this function from our tasklet,
@@ -2232,8 +2242,9 @@ static void kcryptd_queue_crypt(struct dm_crypt_io *io)
{
struct crypt_config *cc = io->cc;
- if ((bio_data_dir(io->base_bio) == READ && test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags)) ||
- (bio_data_dir(io->base_bio) == WRITE && test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))) {
+ if (static_branch_unlikely(&use_tasklet_enabled) &&
+ ((bio_data_dir(io->base_bio) == READ && test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags)) ||
+ (bio_data_dir(io->base_bio) == WRITE && test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags)))) {
/*
* in_hardirq(): Crypto API's skcipher_walk_first() refuses to work in hard IRQ context.
* irqs_disabled(): the kernel may run some IO completion from the idle thread, but
@@ -2746,6 +2757,10 @@ static void crypt_dtr(struct dm_target *ti)
crypt_calculate_pages_per_client();
spin_unlock(&dm_crypt_clients_lock);
+ if (test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags) ||
+ test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))
+ static_branch_dec(&use_tasklet_enabled);
+
dm_audit_log_dtr(DM_MSG_PREFIX, ti, 1);
}
@@ -3375,6 +3390,10 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
ti->limit_swap_bios = true;
ti->accounts_remapped_io = true;
+ if (test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags) ||
+ test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))
+ static_branch_inc(&use_tasklet_enabled);
+
dm_audit_log_ctr(DM_MSG_PREFIX, ti, 1);
return 0;
--
2.37.1 (Apple Git-137.1)
Hi,
On 3/7/2023 10:47 PM, Mike Snitzer wrote:
> On Mon, Mar 06 2023 at 9:12P -0500,
> Hou Tao <[email protected]> wrote:
>
>> Hi,
>>
>> On 3/7/2023 3:31 AM, Mike Snitzer wrote:
>>> On Mon, Mar 06 2023 at 8:49P -0500,
>>> Hou Tao <[email protected]> wrote:
>>>
>>>> From: Hou Tao <[email protected]>
>>>>
>>>> When neither no_read_workqueue nor no_write_workqueue are enabled,
>>>> tasklet_trylock() in crypt_dec_pending() may still return false due to
>>>> an uninitialized state, and dm-crypt will do io completion in io_queue
>>>> instead of current context unnecessarily.
>>> Have you actually experienced this?
>> Yes. I had written a bpftrace script to check the completion context of
>> blkdev_bio_end_io_simple() when doing direct io read on dm-crypt device. The
>> expected context should be unbound workers of crypt_queue, but sometimes the
>> context is the bound worker of io_queue.
> OK, thanks for clarifying. Curious to know the circumstance (I
> thought per-bio-data is zero'd -- but it may be I'm mistaken).
The circumstance is just a normal qemu VM running the vanilla kernel for test
purpose. According to the implementation of bio_alloc_bioset(), the front pad of
bio is not initialized and only bio itself is initialized. AFAIK if
CONFIG_INIT_ON_ALLOC_DEFAULT_ON is enabled, per-bio-data may be zeroed.
>
> I won't be marking this commit for stable@ but if others feel
> differently please let me know and I'll do so. (We can always propose
> it to stable@, after the fact, even if the commit header doesn't Cc
> stable@)
>
>>>> Fix it by initializing io->tasklet in crypt_io_init().
>>> Really would rather avoid always calling tasklet_init(). But I can
>>> optimize it away with a later patch.
>> My first though was "io->tasklet.state = 0", but it may be fragile because it
>> operated on the internal status of tasklet, so I switch to tasklet_init().
> Yes, I looked into it and came up with the same hack.. and I too felt
> it was too fragile due to open-coding direct access to the tasklet's
> members.
>
> I have a patch I just staged that staged that uses jump_labels to
> optimize this code. If you might review/test/verify it works well for
> you that'd be appreciated:
> https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=ae75a25bd83f7c541240449d2fff3a44433e506b
>
> It builds on your patch, which I added a comment to:
> https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=d9fe0a98a2e0a1cf585e8a6555afb33be968bd13
Thanks for the comments. It is fine to me.
>
> From: Mike Snitzer <[email protected]>
> Date: Mon, 6 Mar 2023 15:58:33 -0500
> Subject: [PATCH] dm crypt: conditionally enable code needed for tasklet usecases
>
> Use jump_label to limit the need for branching, and tasklet_init(),
> unless either of the optional "no_read_workqueue" and/or
> "no_write_workqueue" features are used.
>
> Signed-off-by: Mike Snitzer <[email protected]>
> ---
> drivers/md/dm-crypt.c | 35 +++++++++++++++++++++++++++--------
> 1 file changed, 27 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> index 641457e72603..2d0309ca07f5 100644
> --- a/drivers/md/dm-crypt.c
> +++ b/drivers/md/dm-crypt.c
> @@ -40,6 +40,7 @@
> #include <keys/user-type.h>
> #include <keys/encrypted-type.h>
> #include <keys/trusted-type.h>
> +#include <linux/jump_label.h>
>
> #include <linux/device-mapper.h>
>
> @@ -85,6 +86,8 @@ struct dm_crypt_io {
> struct rb_node rb_node;
> } CRYPTO_MINALIGN_ATTR;
>
> +static DEFINE_STATIC_KEY_FALSE(use_tasklet_enabled);
> +
> struct dm_crypt_request {
> struct convert_context *ctx;
> struct scatterlist sg_in[4];
> @@ -1730,12 +1733,15 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
> io->sector = sector;
> io->error = 0;
> io->ctx.r.req = NULL;
> - /*
> - * tasklet_init() here to ensure crypt_dec_pending()'s
> - * tasklet_trylock() doesn't incorrectly return false
> - * even when tasklet isn't in use.
> - */
> - tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
> + if (static_branch_unlikely(&use_tasklet_enabled)) {
> + /*
> + * tasklet_init() here to ensure crypt_dec_pending()'s
> + * tasklet_trylock() doesn't incorrectly return false
> + * even when tasklet isn't in use.
> + */
> + tasklet_init(&io->tasklet, kcryptd_crypt_tasklet,
> + (unsigned long)&io->work);
> + }
> io->integrity_metadata = NULL;
> io->integrity_metadata_from_pool = false;
> atomic_set(&io->io_pending, 0);
> @@ -1775,6 +1781,10 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
> kfree(io->integrity_metadata);
>
> base_bio->bi_status = error;
> + if (!static_branch_unlikely(&use_tasklet_enabled)) {
> + bio_endio(base_bio);
> + return;
> + }
Because use_tasklet_enabled can be enabled concurrently, so I think it is still
possible that crypt_dec_pending will try-lock an unitialized tasklet if
use_tasklet_enabled is enabled when invoking crypt_dec_pending().
>
> /*
> * If we are running this function from our tasklet,
> @@ -2232,8 +2242,9 @@ static void kcryptd_queue_crypt(struct dm_crypt_io *io)
> {
> struct crypt_config *cc = io->cc;
>
> - if ((bio_data_dir(io->base_bio) == READ && test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags)) ||
> - (bio_data_dir(io->base_bio) == WRITE && test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))) {
> + if (static_branch_unlikely(&use_tasklet_enabled) &&
> + ((bio_data_dir(io->base_bio) == READ && test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags)) ||
> + (bio_data_dir(io->base_bio) == WRITE && test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags)))) {
> /*
> * in_hardirq(): Crypto API's skcipher_walk_first() refuses to work in hard IRQ context.
> * irqs_disabled(): the kernel may run some IO completion from the idle thread, but
> @@ -2746,6 +2757,10 @@ static void crypt_dtr(struct dm_target *ti)
> crypt_calculate_pages_per_client();
> spin_unlock(&dm_crypt_clients_lock);
>
> + if (test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags) ||
> + test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))
> + static_branch_dec(&use_tasklet_enabled);
> +
> dm_audit_log_dtr(DM_MSG_PREFIX, ti, 1);
> }
>
> @@ -3375,6 +3390,10 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
> ti->limit_swap_bios = true;
> ti->accounts_remapped_io = true;
>
> + if (test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags) ||
> + test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))
> + static_branch_inc(&use_tasklet_enabled);
> +
> dm_audit_log_ctr(DM_MSG_PREFIX, ti, 1);
> return 0;
>
On Wed, Mar 8, 2023 at 2:56 AM Hou Tao <[email protected]> wrote:
>
> Hi,
>
> On 3/7/2023 10:47 PM, Mike Snitzer wrote:
> > On Mon, Mar 06 2023 at 9:12P -0500,
> > Hou Tao <[email protected]> wrote:
> >
> >> Hi,
> >>
> >> On 3/7/2023 3:31 AM, Mike Snitzer wrote:
> >>> On Mon, Mar 06 2023 at 8:49P -0500,
> >>> Hou Tao <[email protected]> wrote:
> >>>
> >>>> From: Hou Tao <[email protected]>
> >>>>
> >>>> When neither no_read_workqueue nor no_write_workqueue are enabled,
> >>>> tasklet_trylock() in crypt_dec_pending() may still return false due to
> >>>> an uninitialized state, and dm-crypt will do io completion in io_queue
> >>>> instead of current context unnecessarily.
> >>> Have you actually experienced this?
> >> Yes. I had written a bpftrace script to check the completion context of
> >> blkdev_bio_end_io_simple() when doing direct io read on dm-crypt device. The
> >> expected context should be unbound workers of crypt_queue, but sometimes the
> >> context is the bound worker of io_queue.
> > OK, thanks for clarifying. Curious to know the circumstance (I
> > thought per-bio-data is zero'd -- but it may be I'm mistaken).
> The circumstance is just a normal qemu VM running the vanilla kernel for test
> purpose. According to the implementation of bio_alloc_bioset(), the front pad of
> bio is not initialized and only bio itself is initialized. AFAIK if
> CONFIG_INIT_ON_ALLOC_DEFAULT_ON is enabled, per-bio-data may be zeroed.
> >
> > I won't be marking this commit for stable@ but if others feel
> > differently please let me know and I'll do so. (We can always propose
> > it to stable@, after the fact, even if the commit header doesn't Cc
> > stable@)
> >
> >>>> Fix it by initializing io->tasklet in crypt_io_init().
> >>> Really would rather avoid always calling tasklet_init(). But I can
> >>> optimize it away with a later patch.
> >> My first though was "io->tasklet.state = 0", but it may be fragile because it
> >> operated on the internal status of tasklet, so I switch to tasklet_init().
> > Yes, I looked into it and came up with the same hack.. and I too felt
> > it was too fragile due to open-coding direct access to the tasklet's
> > members.
> >
> > I have a patch I just staged that staged that uses jump_labels to
> > optimize this code. If you might review/test/verify it works well for
> > you that'd be appreciated:
> > https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=ae75a25bd83f7c541240449d2fff3a44433e506b
> >
> > It builds on your patch, which I added a comment to:
> > https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=d9fe0a98a2e0a1cf585e8a6555afb33be968bd13
> Thanks for the comments. It is fine to me.
> >
> > From: Mike Snitzer <[email protected]>
> > Date: Mon, 6 Mar 2023 15:58:33 -0500
> > Subject: [PATCH] dm crypt: conditionally enable code needed for tasklet usecases
> >
> > Use jump_label to limit the need for branching, and tasklet_init(),
> > unless either of the optional "no_read_workqueue" and/or
> > "no_write_workqueue" features are used.
> >
> > Signed-off-by: Mike Snitzer <[email protected]>
> > ---
> > drivers/md/dm-crypt.c | 35 +++++++++++++++++++++++++++--------
> > 1 file changed, 27 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> > index 641457e72603..2d0309ca07f5 100644
> > --- a/drivers/md/dm-crypt.c
> > +++ b/drivers/md/dm-crypt.c
> > @@ -40,6 +40,7 @@
> > #include <keys/user-type.h>
> > #include <keys/encrypted-type.h>
> > #include <keys/trusted-type.h>
> > +#include <linux/jump_label.h>
> >
> > #include <linux/device-mapper.h>
> >
> > @@ -85,6 +86,8 @@ struct dm_crypt_io {
> > struct rb_node rb_node;
> > } CRYPTO_MINALIGN_ATTR;
> >
> > +static DEFINE_STATIC_KEY_FALSE(use_tasklet_enabled);
> > +
> > struct dm_crypt_request {
> > struct convert_context *ctx;
> > struct scatterlist sg_in[4];
> > @@ -1730,12 +1733,15 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
> > io->sector = sector;
> > io->error = 0;
> > io->ctx.r.req = NULL;
> > - /*
> > - * tasklet_init() here to ensure crypt_dec_pending()'s
> > - * tasklet_trylock() doesn't incorrectly return false
> > - * even when tasklet isn't in use.
> > - */
> > - tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
> > + if (static_branch_unlikely(&use_tasklet_enabled)) {
> > + /*
> > + * tasklet_init() here to ensure crypt_dec_pending()'s
> > + * tasklet_trylock() doesn't incorrectly return false
> > + * even when tasklet isn't in use.
> > + */
> > + tasklet_init(&io->tasklet, kcryptd_crypt_tasklet,
> > + (unsigned long)&io->work);
> > + }
> > io->integrity_metadata = NULL;
> > io->integrity_metadata_from_pool = false;
> > atomic_set(&io->io_pending, 0);
> > @@ -1775,6 +1781,10 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
> > kfree(io->integrity_metadata);
> >
> > base_bio->bi_status = error;
> > + if (!static_branch_unlikely(&use_tasklet_enabled)) {
> > + bio_endio(base_bio);
> > + return;
> > + }
> Because use_tasklet_enabled can be enabled concurrently, so I think it is still
> possible that crypt_dec_pending will try-lock an unitialized tasklet if
> use_tasklet_enabled is enabled when invoking crypt_dec_pending().
Perhaps instead we can just pass an additional flag from
tasklet_schedule to indicate to the function that we're running in a
tasklet. I originally have chosen the tasklet_trylock/unlock hack to
avoid passing an extra flag. But unitialized memory makes sense as
well as the desire to avoid calling tasklet_init unconditionally. So
an extra member in dm_crypt_io might be the most straightforward here.
Ignat
> > /*
> > * If we are running this function from our tasklet,
> > @@ -2232,8 +2242,9 @@ static void kcryptd_queue_crypt(struct dm_crypt_io *io)
> > {
> > struct crypt_config *cc = io->cc;
> >
> > - if ((bio_data_dir(io->base_bio) == READ && test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags)) ||
> > - (bio_data_dir(io->base_bio) == WRITE && test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))) {
> > + if (static_branch_unlikely(&use_tasklet_enabled) &&
> > + ((bio_data_dir(io->base_bio) == READ && test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags)) ||
> > + (bio_data_dir(io->base_bio) == WRITE && test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags)))) {
> > /*
> > * in_hardirq(): Crypto API's skcipher_walk_first() refuses to work in hard IRQ context.
> > * irqs_disabled(): the kernel may run some IO completion from the idle thread, but
> > @@ -2746,6 +2757,10 @@ static void crypt_dtr(struct dm_target *ti)
> > crypt_calculate_pages_per_client();
> > spin_unlock(&dm_crypt_clients_lock);
> >
> > + if (test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags) ||
> > + test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))
> > + static_branch_dec(&use_tasklet_enabled);
> > +
> > dm_audit_log_dtr(DM_MSG_PREFIX, ti, 1);
> > }
> >
> > @@ -3375,6 +3390,10 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
> > ti->limit_swap_bios = true;
> > ti->accounts_remapped_io = true;
> >
> > + if (test_bit(DM_CRYPT_NO_READ_WORKQUEUE, &cc->flags) ||
> > + test_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags))
> > + static_branch_inc(&use_tasklet_enabled);
> > +
> > dm_audit_log_ctr(DM_MSG_PREFIX, ti, 1);
> > return 0;
> >
>
On Wed, Mar 08 2023 at 8:55P -0500,
Ignat Korchagin <[email protected]> wrote:
> On Wed, Mar 8, 2023 at 2:56 AM Hou Tao <[email protected]> wrote:
> >
> > Hi,
> >
> > On 3/7/2023 10:47 PM, Mike Snitzer wrote:
> > > On Mon, Mar 06 2023 at 9:12P -0500,
> > > Hou Tao <[email protected]> wrote:
> > >
> > >> Hi,
> > >>
> > >> On 3/7/2023 3:31 AM, Mike Snitzer wrote:
> > >>> On Mon, Mar 06 2023 at 8:49P -0500,
> > >>> Hou Tao <[email protected]> wrote:
> > >>>
> > >>>> From: Hou Tao <[email protected]>
> > >>>>
> > >>>> When neither no_read_workqueue nor no_write_workqueue are enabled,
> > >>>> tasklet_trylock() in crypt_dec_pending() may still return false due to
> > >>>> an uninitialized state, and dm-crypt will do io completion in io_queue
> > >>>> instead of current context unnecessarily.
> > >>> Have you actually experienced this?
> > >> Yes. I had written a bpftrace script to check the completion context of
> > >> blkdev_bio_end_io_simple() when doing direct io read on dm-crypt device. The
> > >> expected context should be unbound workers of crypt_queue, but sometimes the
> > >> context is the bound worker of io_queue.
> > > OK, thanks for clarifying. Curious to know the circumstance (I
> > > thought per-bio-data is zero'd -- but it may be I'm mistaken).
> > The circumstance is just a normal qemu VM running the vanilla kernel for test
> > purpose. According to the implementation of bio_alloc_bioset(), the front pad of
> > bio is not initialized and only bio itself is initialized. AFAIK if
> > CONFIG_INIT_ON_ALLOC_DEFAULT_ON is enabled, per-bio-data may be zeroed.
OK.
> > > From: Mike Snitzer <[email protected]>
> > > Date: Mon, 6 Mar 2023 15:58:33 -0500
> > > Subject: [PATCH] dm crypt: conditionally enable code needed for tasklet usecases
> > >
> > > Use jump_label to limit the need for branching, and tasklet_init(),
> > > unless either of the optional "no_read_workqueue" and/or
> > > "no_write_workqueue" features are used.
> > >
> > > Signed-off-by: Mike Snitzer <[email protected]>
> > > ---
> > > drivers/md/dm-crypt.c | 35 +++++++++++++++++++++++++++--------
> > > 1 file changed, 27 insertions(+), 8 deletions(-)
> > >
> > > diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> > > index 641457e72603..2d0309ca07f5 100644
> > > --- a/drivers/md/dm-crypt.c
> > > +++ b/drivers/md/dm-crypt.c
> > > @@ -40,6 +40,7 @@
> > > #include <keys/user-type.h>
> > > #include <keys/encrypted-type.h>
> > > #include <keys/trusted-type.h>
> > > +#include <linux/jump_label.h>
> > >
> > > #include <linux/device-mapper.h>
> > >
> > > @@ -85,6 +86,8 @@ struct dm_crypt_io {
> > > struct rb_node rb_node;
> > > } CRYPTO_MINALIGN_ATTR;
> > >
> > > +static DEFINE_STATIC_KEY_FALSE(use_tasklet_enabled);
> > > +
> > > struct dm_crypt_request {
> > > struct convert_context *ctx;
> > > struct scatterlist sg_in[4];
> > > @@ -1730,12 +1733,15 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
> > > io->sector = sector;
> > > io->error = 0;
> > > io->ctx.r.req = NULL;
> > > - /*
> > > - * tasklet_init() here to ensure crypt_dec_pending()'s
> > > - * tasklet_trylock() doesn't incorrectly return false
> > > - * even when tasklet isn't in use.
> > > - */
> > > - tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
> > > + if (static_branch_unlikely(&use_tasklet_enabled)) {
> > > + /*
> > > + * tasklet_init() here to ensure crypt_dec_pending()'s
> > > + * tasklet_trylock() doesn't incorrectly return false
> > > + * even when tasklet isn't in use.
> > > + */
> > > + tasklet_init(&io->tasklet, kcryptd_crypt_tasklet,
> > > + (unsigned long)&io->work);
> > > + }
> > > io->integrity_metadata = NULL;
> > > io->integrity_metadata_from_pool = false;
> > > atomic_set(&io->io_pending, 0);
> > > @@ -1775,6 +1781,10 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
> > > kfree(io->integrity_metadata);
> > >
> > > base_bio->bi_status = error;
> > > + if (!static_branch_unlikely(&use_tasklet_enabled)) {
> > > + bio_endio(base_bio);
> > > + return;
> > > + }
> > Because use_tasklet_enabled can be enabled concurrently, so I think it is still
> > possible that crypt_dec_pending will try-lock an unitialized tasklet if
> > use_tasklet_enabled is enabled when invoking crypt_dec_pending().
Good point, while I think it is probably acceptable given the worst
case is punting the bio_endio to a workqueue for a time ...
> Perhaps instead we can just pass an additional flag from
> tasklet_schedule to indicate to the function that we're running in a
> tasklet. I originally have chosen the tasklet_trylock/unlock hack to
> avoid passing an extra flag. But unitialized memory makes sense as
> well as the desire to avoid calling tasklet_init unconditionally. So
> an extra member in dm_crypt_io might be the most straightforward here.
... I think we should certainly evaluate the use of an extra flag.
Ignat: I'll have a look at implementing it but if you have a patch
already developed please do share.
Thanks,
Mike
On Wed, Mar 08 2023 at 2:19P -0500,
Mike Snitzer <[email protected]> wrote:
> On Wed, Mar 08 2023 at 8:55P -0500,
> Ignat Korchagin <[email protected]> wrote:
>
> > Perhaps instead we can just pass an additional flag from
> > tasklet_schedule to indicate to the function that we're running in a
> > tasklet. I originally have chosen the tasklet_trylock/unlock hack to
> > avoid passing an extra flag. But unitialized memory makes sense as
> > well as the desire to avoid calling tasklet_init unconditionally. So
> > an extra member in dm_crypt_io might be the most straightforward here.
>
> ... I think we should certainly evaluate the use of an extra flag.
>
> Ignat: I'll have a look at implementing it but if you have a patch
> already developed please do share.
I've staged the following in linux-next for 6.3 via the linux-dm.git,
but if you see anything wrong with it I can obviously fix:
From: Mike Snitzer <[email protected]>
Date: Wed, 8 Mar 2023 14:39:54 -0500
Subject: [PATCH] dm crypt: avoid accessing uninitialized tasklet
When neither "no_read_workqueue" nor "no_write_workqueue" are enabled,
tasklet_trylock() in crypt_dec_pending() may still return false due to
an uninitialized state, and dm-crypt will unnecessarily do io completion
in io_queue workqueue instead of current context.
Fix this by adding an 'in_tasklet' flag to dm_crypt_io struct and
initialize it to false in crypt_io_init(). Set this flag to true in
kcryptd_queue_crypt() before calling tasklet_schedule(). If set
crypt_dec_pending() will punt io completion to a workqueue.
This also nicely avoids the tasklet_trylock/unlock hack when tasklets
aren't in use.
Fixes: 8e14f610159d ("dm crypt: do not call bio_endio() from the dm-crypt tasklet")
Cc: [email protected]
Reported-by: Hou Tao <[email protected]>
Suggested-by: Ignat Korchagin <[email protected]>
Signed-off-by: Mike Snitzer <[email protected]>
---
drivers/md/dm-crypt.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index faba1be572f9..de08ff4f7c98 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -72,7 +72,9 @@ struct dm_crypt_io {
struct crypt_config *cc;
struct bio *base_bio;
u8 *integrity_metadata;
- bool integrity_metadata_from_pool;
+ bool integrity_metadata_from_pool:1;
+ bool in_tasklet:1;
+
struct work_struct work;
struct tasklet_struct tasklet;
@@ -1731,6 +1733,7 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
io->ctx.r.req = NULL;
io->integrity_metadata = NULL;
io->integrity_metadata_from_pool = false;
+ io->in_tasklet = false;
atomic_set(&io->io_pending, 0);
}
@@ -1777,8 +1780,7 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
* our tasklet. In this case we need to delay bio_endio()
* execution to after the tasklet is done and dequeued.
*/
- if (tasklet_trylock(&io->tasklet)) {
- tasklet_unlock(&io->tasklet);
+ if (!io->in_tasklet) {
bio_endio(base_bio);
return;
}
@@ -2233,6 +2235,7 @@ static void kcryptd_queue_crypt(struct dm_crypt_io *io)
* it is being executed with irqs disabled.
*/
if (in_hardirq() || irqs_disabled()) {
+ io->in_tasklet = true;
tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
tasklet_schedule(&io->tasklet);
return;
--
2.37.1 (Apple Git-137.1)
On Wed, Mar 8, 2023 at 7:19 PM Mike Snitzer <[email protected]> wrote:
>
> On Wed, Mar 08 2023 at 8:55P -0500,
> Ignat Korchagin <[email protected]> wrote:
>
> > On Wed, Mar 8, 2023 at 2:56 AM Hou Tao <[email protected]> wrote:
> > >
> > > Hi,
> > >
> > > On 3/7/2023 10:47 PM, Mike Snitzer wrote:
> > > > On Mon, Mar 06 2023 at 9:12P -0500,
> > > > Hou Tao <[email protected]> wrote:
> > > >
> > > >> Hi,
> > > >>
> > > >> On 3/7/2023 3:31 AM, Mike Snitzer wrote:
> > > >>> On Mon, Mar 06 2023 at 8:49P -0500,
> > > >>> Hou Tao <[email protected]> wrote:
> > > >>>
> > > >>>> From: Hou Tao <[email protected]>
> > > >>>>
> > > >>>> When neither no_read_workqueue nor no_write_workqueue are enabled,
> > > >>>> tasklet_trylock() in crypt_dec_pending() may still return false due to
> > > >>>> an uninitialized state, and dm-crypt will do io completion in io_queue
> > > >>>> instead of current context unnecessarily.
> > > >>> Have you actually experienced this?
> > > >> Yes. I had written a bpftrace script to check the completion context of
> > > >> blkdev_bio_end_io_simple() when doing direct io read on dm-crypt device. The
> > > >> expected context should be unbound workers of crypt_queue, but sometimes the
> > > >> context is the bound worker of io_queue.
> > > > OK, thanks for clarifying. Curious to know the circumstance (I
> > > > thought per-bio-data is zero'd -- but it may be I'm mistaken).
> > > The circumstance is just a normal qemu VM running the vanilla kernel for test
> > > purpose. According to the implementation of bio_alloc_bioset(), the front pad of
> > > bio is not initialized and only bio itself is initialized. AFAIK if
> > > CONFIG_INIT_ON_ALLOC_DEFAULT_ON is enabled, per-bio-data may be zeroed.
>
> OK.
>
> > > > From: Mike Snitzer <[email protected]>
> > > > Date: Mon, 6 Mar 2023 15:58:33 -0500
> > > > Subject: [PATCH] dm crypt: conditionally enable code needed for tasklet usecases
> > > >
> > > > Use jump_label to limit the need for branching, and tasklet_init(),
> > > > unless either of the optional "no_read_workqueue" and/or
> > > > "no_write_workqueue" features are used.
> > > >
> > > > Signed-off-by: Mike Snitzer <[email protected]>
> > > > ---
> > > > drivers/md/dm-crypt.c | 35 +++++++++++++++++++++++++++--------
> > > > 1 file changed, 27 insertions(+), 8 deletions(-)
> > > >
> > > > diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> > > > index 641457e72603..2d0309ca07f5 100644
> > > > --- a/drivers/md/dm-crypt.c
> > > > +++ b/drivers/md/dm-crypt.c
> > > > @@ -40,6 +40,7 @@
> > > > #include <keys/user-type.h>
> > > > #include <keys/encrypted-type.h>
> > > > #include <keys/trusted-type.h>
> > > > +#include <linux/jump_label.h>
> > > >
> > > > #include <linux/device-mapper.h>
> > > >
> > > > @@ -85,6 +86,8 @@ struct dm_crypt_io {
> > > > struct rb_node rb_node;
> > > > } CRYPTO_MINALIGN_ATTR;
> > > >
> > > > +static DEFINE_STATIC_KEY_FALSE(use_tasklet_enabled);
> > > > +
> > > > struct dm_crypt_request {
> > > > struct convert_context *ctx;
> > > > struct scatterlist sg_in[4];
> > > > @@ -1730,12 +1733,15 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
> > > > io->sector = sector;
> > > > io->error = 0;
> > > > io->ctx.r.req = NULL;
> > > > - /*
> > > > - * tasklet_init() here to ensure crypt_dec_pending()'s
> > > > - * tasklet_trylock() doesn't incorrectly return false
> > > > - * even when tasklet isn't in use.
> > > > - */
> > > > - tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
> > > > + if (static_branch_unlikely(&use_tasklet_enabled)) {
> > > > + /*
> > > > + * tasklet_init() here to ensure crypt_dec_pending()'s
> > > > + * tasklet_trylock() doesn't incorrectly return false
> > > > + * even when tasklet isn't in use.
> > > > + */
> > > > + tasklet_init(&io->tasklet, kcryptd_crypt_tasklet,
> > > > + (unsigned long)&io->work);
> > > > + }
> > > > io->integrity_metadata = NULL;
> > > > io->integrity_metadata_from_pool = false;
> > > > atomic_set(&io->io_pending, 0);
> > > > @@ -1775,6 +1781,10 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
> > > > kfree(io->integrity_metadata);
> > > >
> > > > base_bio->bi_status = error;
> > > > + if (!static_branch_unlikely(&use_tasklet_enabled)) {
> > > > + bio_endio(base_bio);
> > > > + return;
> > > > + }
> > > Because use_tasklet_enabled can be enabled concurrently, so I think it is still
> > > possible that crypt_dec_pending will try-lock an unitialized tasklet if
> > > use_tasklet_enabled is enabled when invoking crypt_dec_pending().
>
> Good point, while I think it is probably acceptable given the worst
> case is punting the bio_endio to a workqueue for a time ...
>
> > Perhaps instead we can just pass an additional flag from
> > tasklet_schedule to indicate to the function that we're running in a
> > tasklet. I originally have chosen the tasklet_trylock/unlock hack to
> > avoid passing an extra flag. But unitialized memory makes sense as
> > well as the desire to avoid calling tasklet_init unconditionally. So
> > an extra member in dm_crypt_io might be the most straightforward here.
>
> ... I think we should certainly evaluate the use of an extra flag.
>
> Ignat: I'll have a look at implementing it but if you have a patch
> already developed please do share.
I don't have but it seems your latest patch is exactly what I had in mind.
> Thanks,
> Mike
On Wed, Mar 8, 2023 at 8:27 PM Mike Snitzer <[email protected]> wrote:
>
> On Wed, Mar 08 2023 at 2:19P -0500,
> Mike Snitzer <[email protected]> wrote:
>
> > On Wed, Mar 08 2023 at 8:55P -0500,
> > Ignat Korchagin <[email protected]> wrote:
> >
> > > Perhaps instead we can just pass an additional flag from
> > > tasklet_schedule to indicate to the function that we're running in a
> > > tasklet. I originally have chosen the tasklet_trylock/unlock hack to
> > > avoid passing an extra flag. But unitialized memory makes sense as
> > > well as the desire to avoid calling tasklet_init unconditionally. So
> > > an extra member in dm_crypt_io might be the most straightforward here.
> >
> > ... I think we should certainly evaluate the use of an extra flag.
> >
> > Ignat: I'll have a look at implementing it but if you have a patch
> > already developed please do share.
>
> I've staged the following in linux-next for 6.3 via the linux-dm.git,
> but if you see anything wrong with it I can obviously fix:
>
> From: Mike Snitzer <[email protected]>
> Date: Wed, 8 Mar 2023 14:39:54 -0500
> Subject: [PATCH] dm crypt: avoid accessing uninitialized tasklet
>
> When neither "no_read_workqueue" nor "no_write_workqueue" are enabled,
> tasklet_trylock() in crypt_dec_pending() may still return false due to
> an uninitialized state, and dm-crypt will unnecessarily do io completion
> in io_queue workqueue instead of current context.
>
> Fix this by adding an 'in_tasklet' flag to dm_crypt_io struct and
> initialize it to false in crypt_io_init(). Set this flag to true in
> kcryptd_queue_crypt() before calling tasklet_schedule(). If set
> crypt_dec_pending() will punt io completion to a workqueue.
>
> This also nicely avoids the tasklet_trylock/unlock hack when tasklets
> aren't in use.
>
> Fixes: 8e14f610159d ("dm crypt: do not call bio_endio() from the dm-crypt tasklet")
> Cc: [email protected]
> Reported-by: Hou Tao <[email protected]>
> Suggested-by: Ignat Korchagin <[email protected]>
> Signed-off-by: Mike Snitzer <[email protected]>
> ---
> drivers/md/dm-crypt.c | 9 ++++++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
> index faba1be572f9..de08ff4f7c98 100644
> --- a/drivers/md/dm-crypt.c
> +++ b/drivers/md/dm-crypt.c
> @@ -72,7 +72,9 @@ struct dm_crypt_io {
> struct crypt_config *cc;
> struct bio *base_bio;
> u8 *integrity_metadata;
> - bool integrity_metadata_from_pool;
> + bool integrity_metadata_from_pool:1;
> + bool in_tasklet:1;
> +
> struct work_struct work;
> struct tasklet_struct tasklet;
>
> @@ -1731,6 +1733,7 @@ static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
> io->ctx.r.req = NULL;
> io->integrity_metadata = NULL;
> io->integrity_metadata_from_pool = false;
> + io->in_tasklet = false;
> atomic_set(&io->io_pending, 0);
> }
>
> @@ -1777,8 +1780,7 @@ static void crypt_dec_pending(struct dm_crypt_io *io)
> * our tasklet. In this case we need to delay bio_endio()
> * execution to after the tasklet is done and dequeued.
> */
> - if (tasklet_trylock(&io->tasklet)) {
> - tasklet_unlock(&io->tasklet);
> + if (!io->in_tasklet) {
nitpick: maybe invert the logic here for better readability? (so it
becomes "if (in_tasklet) queue..." else just falls through bio_endio()
)
> bio_endio(base_bio);
> return;
> }
> @@ -2233,6 +2235,7 @@ static void kcryptd_queue_crypt(struct dm_crypt_io *io)
> * it is being executed with irqs disabled.
> */
> if (in_hardirq() || irqs_disabled()) {
> + io->in_tasklet = true;
> tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
> tasklet_schedule(&io->tasklet);
> return;
> --
> 2.37.1 (Apple Git-137.1)
>
Reviewed-by: Ignat Korchagin <[email protected]>