2019-01-30 10:30:21

by Javier González

[permalink] [raw]
Subject: [PATCH V2] lightnvm: pblk: prevent stall due to wb threshold

In order to respect mw_cuinits, pblk's write buffer maintains a
backpointer to protect data not yet persisted; when writing to the write
buffer, this backpointer defines a threshold that pblk's rate-limiter
enforces.

On small PU configurations, the following scenarios might take place: (i)
the threshold is larger than the write buffer and (ii) the threshold is
smaller than the write buffer, but larger than the maximun allowed
split bio - 256KB at this moment (Note that writes are not always
split - we only do this when we the size of the buffer is smaller
than the buffer). In both cases, pblk's rate-limiter prevents the I/O to
be written to the buffer, thus stalling.

This patch fixes the original backpointer implementation by considering
the threshold both on buffer creation and on the rate-limiters path,
when bio_split is triggered (case (ii) above).

Fixes: 766c8ceb16fc ("lightnvm: pblk: guarantee that backpointer is respected on writer stall")
Signed-off-by: Javier González <[email protected]>
---

Changes since V1:
- Fix a bad arithmetinc on the rate-limiter max_io calculation (from
Hans)

drivers/lightnvm/pblk-rb.c | 25 +++++++++++++++++++------
drivers/lightnvm/pblk-rl.c | 5 ++---
drivers/lightnvm/pblk.h | 2 +-
3 files changed, 22 insertions(+), 10 deletions(-)

diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c
index d4ca8c64ee0f..a6133b50ed9c 100644
--- a/drivers/lightnvm/pblk-rb.c
+++ b/drivers/lightnvm/pblk-rb.c
@@ -45,10 +45,23 @@ void pblk_rb_free(struct pblk_rb *rb)
/*
* pblk_rb_calculate_size -- calculate the size of the write buffer
*/
-static unsigned int pblk_rb_calculate_size(unsigned int nr_entries)
+static unsigned int pblk_rb_calculate_size(unsigned int nr_entries,
+ unsigned int threshold)
{
- /* Alloc a write buffer that can at least fit 128 entries */
- return (1 << max(get_count_order(nr_entries), 7));
+ unsigned int thr_sz = 1 << (get_count_order(threshold + NVM_MAX_VLBA));
+ unsigned int max_sz = max(thr_sz, nr_entries);
+ unsigned int max_io;
+
+ /* Alloc a write buffer that can (i) fit at least two split bios
+ * (considering max I/O size NVM_MAX_VLBA, and (ii) guarantee that the
+ * threshold will be respected
+ */
+ max_io = (1 << max((int)(get_count_order(max_sz)),
+ (int)(get_count_order(NVM_MAX_VLBA << 1))));
+ if ((threshold + NVM_MAX_VLBA) >= max_io)
+ max_io <<= 1;
+
+ return max_io;
}

/*
@@ -67,12 +80,12 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
unsigned int alloc_order, order, iter;
unsigned int nr_entries;

- nr_entries = pblk_rb_calculate_size(size);
+ nr_entries = pblk_rb_calculate_size(size, threshold);
entries = vzalloc(array_size(nr_entries, sizeof(struct pblk_rb_entry)));
if (!entries)
return -ENOMEM;

- power_size = get_count_order(size);
+ power_size = get_count_order(nr_entries);
power_seg_sz = get_count_order(seg_size);

down_write(&pblk_rb_lock);
@@ -149,7 +162,7 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
* Initialize rate-limiter, which controls access to the write buffer
* by user and GC I/O
*/
- pblk_rl_init(&pblk->rl, rb->nr_entries);
+ pblk_rl_init(&pblk->rl, rb->nr_entries, threshold);

return 0;
}
diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
index 76116d5f78e4..e9e0af0df165 100644
--- a/drivers/lightnvm/pblk-rl.c
+++ b/drivers/lightnvm/pblk-rl.c
@@ -207,7 +207,7 @@ void pblk_rl_free(struct pblk_rl *rl)
del_timer(&rl->u_timer);
}

-void pblk_rl_init(struct pblk_rl *rl, int budget)
+void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold)
{
struct pblk *pblk = container_of(rl, struct pblk, rl);
struct nvm_tgt_dev *dev = pblk->dev;
@@ -217,7 +217,6 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
int sec_meta, blk_meta;
unsigned int rb_windows;

-
/* Consider sectors used for metadata */
sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
@@ -234,7 +233,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
/* To start with, all buffer is available to user I/O writers */
rl->rb_budget = budget;
rl->rb_user_max = budget;
- rl->rb_max_io = budget >> 1;
+ rl->rb_max_io = budget - threshold;
rl->rb_gc_max = 0;
rl->rb_state = PBLK_RL_HIGH;

diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
index 72ae8755764e..a6386d5acd73 100644
--- a/drivers/lightnvm/pblk.h
+++ b/drivers/lightnvm/pblk.h
@@ -924,7 +924,7 @@ int pblk_gc_sysfs_force(struct pblk *pblk, int force);
/*
* pblk rate limiter
*/
-void pblk_rl_init(struct pblk_rl *rl, int budget);
+void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold);
void pblk_rl_free(struct pblk_rl *rl);
void pblk_rl_update_rates(struct pblk_rl *rl);
int pblk_rl_high_thrs(struct pblk_rl *rl);
--
2.17.1



2019-01-30 13:39:25

by Matias Bjørling

[permalink] [raw]
Subject: Re: [PATCH V2] lightnvm: pblk: prevent stall due to wb threshold

On 1/30/19 11:26 AM, Javier González wrote:
> In order to respect mw_cuinits, pblk's write buffer maintains a
> backpointer to protect data not yet persisted; when writing to the write
> buffer, this backpointer defines a threshold that pblk's rate-limiter
> enforces.
>
> On small PU configurations, the following scenarios might take place: (i)
> the threshold is larger than the write buffer and (ii) the threshold is
> smaller than the write buffer, but larger than the maximun allowed
> split bio - 256KB at this moment (Note that writes are not always
> split - we only do this when we the size of the buffer is smaller
> than the buffer). In both cases, pblk's rate-limiter prevents the I/O to
> be written to the buffer, thus stalling.
>
> This patch fixes the original backpointer implementation by considering
> the threshold both on buffer creation and on the rate-limiters path,
> when bio_split is triggered (case (ii) above).
>
> Fixes: 766c8ceb16fc ("lightnvm: pblk: guarantee that backpointer is respected on writer stall")
> Signed-off-by: Javier González <[email protected]>
> ---
>
> Changes since V1:
> - Fix a bad arithmetinc on the rate-limiter max_io calculation (from
> Hans)
>
> drivers/lightnvm/pblk-rb.c | 25 +++++++++++++++++++------
> drivers/lightnvm/pblk-rl.c | 5 ++---
> drivers/lightnvm/pblk.h | 2 +-
> 3 files changed, 22 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c
> index d4ca8c64ee0f..a6133b50ed9c 100644
> --- a/drivers/lightnvm/pblk-rb.c
> +++ b/drivers/lightnvm/pblk-rb.c
> @@ -45,10 +45,23 @@ void pblk_rb_free(struct pblk_rb *rb)
> /*
> * pblk_rb_calculate_size -- calculate the size of the write buffer
> */
> -static unsigned int pblk_rb_calculate_size(unsigned int nr_entries)
> +static unsigned int pblk_rb_calculate_size(unsigned int nr_entries,
> + unsigned int threshold)
> {
> - /* Alloc a write buffer that can at least fit 128 entries */
> - return (1 << max(get_count_order(nr_entries), 7));
> + unsigned int thr_sz = 1 << (get_count_order(threshold + NVM_MAX_VLBA));
> + unsigned int max_sz = max(thr_sz, nr_entries);
> + unsigned int max_io;
> +
> + /* Alloc a write buffer that can (i) fit at least two split bios
> + * (considering max I/O size NVM_MAX_VLBA, and (ii) guarantee that the
> + * threshold will be respected
> + */
> + max_io = (1 << max((int)(get_count_order(max_sz)),
> + (int)(get_count_order(NVM_MAX_VLBA << 1))));
> + if ((threshold + NVM_MAX_VLBA) >= max_io)
> + max_io <<= 1;
> +
> + return max_io;
> }
>
> /*
> @@ -67,12 +80,12 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> unsigned int alloc_order, order, iter;
> unsigned int nr_entries;
>
> - nr_entries = pblk_rb_calculate_size(size);
> + nr_entries = pblk_rb_calculate_size(size, threshold);
> entries = vzalloc(array_size(nr_entries, sizeof(struct pblk_rb_entry)));
> if (!entries)
> return -ENOMEM;
>
> - power_size = get_count_order(size);
> + power_size = get_count_order(nr_entries);
> power_seg_sz = get_count_order(seg_size);
>
> down_write(&pblk_rb_lock);
> @@ -149,7 +162,7 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> * Initialize rate-limiter, which controls access to the write buffer
> * by user and GC I/O
> */
> - pblk_rl_init(&pblk->rl, rb->nr_entries);
> + pblk_rl_init(&pblk->rl, rb->nr_entries, threshold);
>
> return 0;
> }
> diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
> index 76116d5f78e4..e9e0af0df165 100644
> --- a/drivers/lightnvm/pblk-rl.c
> +++ b/drivers/lightnvm/pblk-rl.c
> @@ -207,7 +207,7 @@ void pblk_rl_free(struct pblk_rl *rl)
> del_timer(&rl->u_timer);
> }
>
> -void pblk_rl_init(struct pblk_rl *rl, int budget)
> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold)
> {
> struct pblk *pblk = container_of(rl, struct pblk, rl);
> struct nvm_tgt_dev *dev = pblk->dev;
> @@ -217,7 +217,6 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> int sec_meta, blk_meta;
> unsigned int rb_windows;
>
> -
> /* Consider sectors used for metadata */
> sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
> blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
> @@ -234,7 +233,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> /* To start with, all buffer is available to user I/O writers */
> rl->rb_budget = budget;
> rl->rb_user_max = budget;
> - rl->rb_max_io = budget >> 1;
> + rl->rb_max_io = budget - threshold;
> rl->rb_gc_max = 0;
> rl->rb_state = PBLK_RL_HIGH;
>
> diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
> index 72ae8755764e..a6386d5acd73 100644
> --- a/drivers/lightnvm/pblk.h
> +++ b/drivers/lightnvm/pblk.h
> @@ -924,7 +924,7 @@ int pblk_gc_sysfs_force(struct pblk *pblk, int force);
> /*
> * pblk rate limiter
> */
> -void pblk_rl_init(struct pblk_rl *rl, int budget);
> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold);
> void pblk_rl_free(struct pblk_rl *rl);
> void pblk_rl_update_rates(struct pblk_rl *rl);
> int pblk_rl_high_thrs(struct pblk_rl *rl);
>

Thanks Javier. Applied for 5.1.

2019-01-31 11:50:36

by Hans Holmberg

[permalink] [raw]
Subject: Re: [PATCH V2] lightnvm: pblk: prevent stall due to wb threshold

Hi Javier!

How did you test this? I'm trying to add a test case to our testing framework.

This is what i ran in qemu, and I got a hang (with this version of the patch)

nvme lnvm create -d nvme0n1 -t pblk -n pblk0 -f -b 0 -e 0

kernel log: [ 116.381799] pblk pblk0: luns:1, lines:280, secs:212736,
buf entries:128

# dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=4k count=1
1+0 records in
1+0 records out
4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000480941 s, 8.5 MB/s
# dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=64k count=1
1+0 records in
1+0 records out
65536 bytes (66 kB, 64 KiB) copied, 0.000477373 s, 137 MB/s
# dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=128k count=1
1+0 records in
1+0 records out
131072 bytes (131 kB, 128 KiB) copied, 0.000548722 s, 239 MB/s
# dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=256k count=1
1+0 records in
1+0 records out
262144 bytes (262 kB, 256 KiB) copied, 0.000718515 s, 365 MB/s
# dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=512k count=1
<HANG>


On Wed, Jan 30, 2019 at 11:28 AM Javier González <[email protected]> wrote:
>
> In order to respect mw_cuinits, pblk's write buffer maintains a
> backpointer to protect data not yet persisted; when writing to the write
> buffer, this backpointer defines a threshold that pblk's rate-limiter
> enforces.
>
> On small PU configurations, the following scenarios might take place: (i)
> the threshold is larger than the write buffer and (ii) the threshold is
> smaller than the write buffer, but larger than the maximun allowed
> split bio - 256KB at this moment (Note that writes are not always
> split - we only do this when we the size of the buffer is smaller
> than the buffer). In both cases, pblk's rate-limiter prevents the I/O to
> be written to the buffer, thus stalling.
>
> This patch fixes the original backpointer implementation by considering
> the threshold both on buffer creation and on the rate-limiters path,
> when bio_split is triggered (case (ii) above).
>
> Fixes: 766c8ceb16fc ("lightnvm: pblk: guarantee that backpointer is respected on writer stall")
> Signed-off-by: Javier González <[email protected]>
> ---
>
> Changes since V1:
> - Fix a bad arithmetinc on the rate-limiter max_io calculation (from
> Hans)
>
> drivers/lightnvm/pblk-rb.c | 25 +++++++++++++++++++------
> drivers/lightnvm/pblk-rl.c | 5 ++---
> drivers/lightnvm/pblk.h | 2 +-
> 3 files changed, 22 insertions(+), 10 deletions(-)
>
> diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c
> index d4ca8c64ee0f..a6133b50ed9c 100644
> --- a/drivers/lightnvm/pblk-rb.c
> +++ b/drivers/lightnvm/pblk-rb.c
> @@ -45,10 +45,23 @@ void pblk_rb_free(struct pblk_rb *rb)
> /*
> * pblk_rb_calculate_size -- calculate the size of the write buffer
> */
> -static unsigned int pblk_rb_calculate_size(unsigned int nr_entries)
> +static unsigned int pblk_rb_calculate_size(unsigned int nr_entries,
> + unsigned int threshold)
> {
> - /* Alloc a write buffer that can at least fit 128 entries */
> - return (1 << max(get_count_order(nr_entries), 7));
> + unsigned int thr_sz = 1 << (get_count_order(threshold + NVM_MAX_VLBA));
> + unsigned int max_sz = max(thr_sz, nr_entries);
> + unsigned int max_io;
> +
> + /* Alloc a write buffer that can (i) fit at least two split bios
> + * (considering max I/O size NVM_MAX_VLBA, and (ii) guarantee that the
> + * threshold will be respected
> + */
> + max_io = (1 << max((int)(get_count_order(max_sz)),
> + (int)(get_count_order(NVM_MAX_VLBA << 1))));
> + if ((threshold + NVM_MAX_VLBA) >= max_io)
> + max_io <<= 1;
> +
> + return max_io;
> }
>
> /*
> @@ -67,12 +80,12 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> unsigned int alloc_order, order, iter;
> unsigned int nr_entries;
>
> - nr_entries = pblk_rb_calculate_size(size);
> + nr_entries = pblk_rb_calculate_size(size, threshold);
> entries = vzalloc(array_size(nr_entries, sizeof(struct pblk_rb_entry)));
> if (!entries)
> return -ENOMEM;
>
> - power_size = get_count_order(size);
> + power_size = get_count_order(nr_entries);
> power_seg_sz = get_count_order(seg_size);
>
> down_write(&pblk_rb_lock);
> @@ -149,7 +162,7 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> * Initialize rate-limiter, which controls access to the write buffer
> * by user and GC I/O
> */
> - pblk_rl_init(&pblk->rl, rb->nr_entries);
> + pblk_rl_init(&pblk->rl, rb->nr_entries, threshold);
>
> return 0;
> }
> diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
> index 76116d5f78e4..e9e0af0df165 100644
> --- a/drivers/lightnvm/pblk-rl.c
> +++ b/drivers/lightnvm/pblk-rl.c
> @@ -207,7 +207,7 @@ void pblk_rl_free(struct pblk_rl *rl)
> del_timer(&rl->u_timer);
> }
>
> -void pblk_rl_init(struct pblk_rl *rl, int budget)
> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold)
> {
> struct pblk *pblk = container_of(rl, struct pblk, rl);
> struct nvm_tgt_dev *dev = pblk->dev;
> @@ -217,7 +217,6 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> int sec_meta, blk_meta;
> unsigned int rb_windows;
>
> -
> /* Consider sectors used for metadata */
> sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
> blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
> @@ -234,7 +233,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> /* To start with, all buffer is available to user I/O writers */
> rl->rb_budget = budget;
> rl->rb_user_max = budget;
> - rl->rb_max_io = budget >> 1;
> + rl->rb_max_io = budget - threshold;
> rl->rb_gc_max = 0;
> rl->rb_state = PBLK_RL_HIGH;
>
> diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
> index 72ae8755764e..a6386d5acd73 100644
> --- a/drivers/lightnvm/pblk.h
> +++ b/drivers/lightnvm/pblk.h
> @@ -924,7 +924,7 @@ int pblk_gc_sysfs_force(struct pblk *pblk, int force);
> /*
> * pblk rate limiter
> */
> -void pblk_rl_init(struct pblk_rl *rl, int budget);
> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold);
> void pblk_rl_free(struct pblk_rl *rl);
> void pblk_rl_update_rates(struct pblk_rl *rl);
> int pblk_rl_high_thrs(struct pblk_rl *rl);
> --
> 2.17.1
>

2019-01-31 16:34:24

by Javier González

[permalink] [raw]
Subject: Re: [PATCH V2] lightnvm: pblk: prevent stall due to wb threshold



> On 31 Jan 2019, at 11.41, Hans Holmberg <[email protected]> wrote:
>
> Hi Javier!
>
> How did you test this? I'm trying to add a test case to our testing framework.
>
> This is what i ran in qemu, and I got a hang (with this version of the patch)
>
> nvme lnvm create -d nvme0n1 -t pblk -n pblk0 -f -b 0 -e 0

I run several low configurations without problem. Can you share the qemu configuration and version?

I’m on travel until Friday - I’ll come back to you over the weekend.

>
> kernel log: [ 116.381799] pblk pblk0: luns:1, lines:280, secs:212736,
> buf entries:128
>
> # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=4k count=1
> 1+0 records in
> 1+0 records out
> 4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000480941 s, 8.5 MB/s
> # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=64k count=1
> 1+0 records in
> 1+0 records out
> 65536 bytes (66 kB, 64 KiB) copied, 0.000477373 s, 137 MB/s
> # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=128k count=1
> 1+0 records in
> 1+0 records out
> 131072 bytes (131 kB, 128 KiB) copied, 0.000548722 s, 239 MB/s
> # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=256k count=1
> 1+0 records in
> 1+0 records out
> 262144 bytes (262 kB, 256 KiB) copied, 0.000718515 s, 365 MB/s
> # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=512k count=1
> <HANG>


>
>
>> On Wed, Jan 30, 2019 at 11:28 AM Javier González <[email protected]> wrote:
>>
>> In order to respect mw_cuinits, pblk's write buffer maintains a
>> backpointer to protect data not yet persisted; when writing to the write
>> buffer, this backpointer defines a threshold that pblk's rate-limiter
>> enforces.
>>
>> On small PU configurations, the following scenarios might take place: (i)
>> the threshold is larger than the write buffer and (ii) the threshold is
>> smaller than the write buffer, but larger than the maximun allowed
>> split bio - 256KB at this moment (Note that writes are not always
>> split - we only do this when we the size of the buffer is smaller
>> than the buffer). In both cases, pblk's rate-limiter prevents the I/O to
>> be written to the buffer, thus stalling.
>>
>> This patch fixes the original backpointer implementation by considering
>> the threshold both on buffer creation and on the rate-limiters path,
>> when bio_split is triggered (case (ii) above).
>>
>> Fixes: 766c8ceb16fc ("lightnvm: pblk: guarantee that backpointer is respected on writer stall")
>> Signed-off-by: Javier González <[email protected]>
>> ---
>>
>> Changes since V1:
>> - Fix a bad arithmetinc on the rate-limiter max_io calculation (from
>> Hans)
>>
>> drivers/lightnvm/pblk-rb.c | 25 +++++++++++++++++++------
>> drivers/lightnvm/pblk-rl.c | 5 ++---
>> drivers/lightnvm/pblk.h | 2 +-
>> 3 files changed, 22 insertions(+), 10 deletions(-)
>>
>> diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c
>> index d4ca8c64ee0f..a6133b50ed9c 100644
>> --- a/drivers/lightnvm/pblk-rb.c
>> +++ b/drivers/lightnvm/pblk-rb.c
>> @@ -45,10 +45,23 @@ void pblk_rb_free(struct pblk_rb *rb)
>> /*
>> * pblk_rb_calculate_size -- calculate the size of the write buffer
>> */
>> -static unsigned int pblk_rb_calculate_size(unsigned int nr_entries)
>> +static unsigned int pblk_rb_calculate_size(unsigned int nr_entries,
>> + unsigned int threshold)
>> {
>> - /* Alloc a write buffer that can at least fit 128 entries */
>> - return (1 << max(get_count_order(nr_entries), 7));
>> + unsigned int thr_sz = 1 << (get_count_order(threshold + NVM_MAX_VLBA));
>> + unsigned int max_sz = max(thr_sz, nr_entries);
>> + unsigned int max_io;
>> +
>> + /* Alloc a write buffer that can (i) fit at least two split bios
>> + * (considering max I/O size NVM_MAX_VLBA, and (ii) guarantee that the
>> + * threshold will be respected
>> + */
>> + max_io = (1 << max((int)(get_count_order(max_sz)),
>> + (int)(get_count_order(NVM_MAX_VLBA << 1))));
>> + if ((threshold + NVM_MAX_VLBA) >= max_io)
>> + max_io <<= 1;
>> +
>> + return max_io;
>> }
>>
>> /*
>> @@ -67,12 +80,12 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
>> unsigned int alloc_order, order, iter;
>> unsigned int nr_entries;
>>
>> - nr_entries = pblk_rb_calculate_size(size);
>> + nr_entries = pblk_rb_calculate_size(size, threshold);
>> entries = vzalloc(array_size(nr_entries, sizeof(struct pblk_rb_entry)));
>> if (!entries)
>> return -ENOMEM;
>>
>> - power_size = get_count_order(size);
>> + power_size = get_count_order(nr_entries);
>> power_seg_sz = get_count_order(seg_size);
>>
>> down_write(&pblk_rb_lock);
>> @@ -149,7 +162,7 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
>> * Initialize rate-limiter, which controls access to the write buffer
>> * by user and GC I/O
>> */
>> - pblk_rl_init(&pblk->rl, rb->nr_entries);
>> + pblk_rl_init(&pblk->rl, rb->nr_entries, threshold);
>>
>> return 0;
>> }
>> diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
>> index 76116d5f78e4..e9e0af0df165 100644
>> --- a/drivers/lightnvm/pblk-rl.c
>> +++ b/drivers/lightnvm/pblk-rl.c
>> @@ -207,7 +207,7 @@ void pblk_rl_free(struct pblk_rl *rl)
>> del_timer(&rl->u_timer);
>> }
>>
>> -void pblk_rl_init(struct pblk_rl *rl, int budget)
>> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold)
>> {
>> struct pblk *pblk = container_of(rl, struct pblk, rl);
>> struct nvm_tgt_dev *dev = pblk->dev;
>> @@ -217,7 +217,6 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
>> int sec_meta, blk_meta;
>> unsigned int rb_windows;
>>
>> -
>> /* Consider sectors used for metadata */
>> sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
>> blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
>> @@ -234,7 +233,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
>> /* To start with, all buffer is available to user I/O writers */
>> rl->rb_budget = budget;
>> rl->rb_user_max = budget;
>> - rl->rb_max_io = budget >> 1;
>> + rl->rb_max_io = budget - threshold;
>> rl->rb_gc_max = 0;
>> rl->rb_state = PBLK_RL_HIGH;
>>
>> diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
>> index 72ae8755764e..a6386d5acd73 100644
>> --- a/drivers/lightnvm/pblk.h
>> +++ b/drivers/lightnvm/pblk.h
>> @@ -924,7 +924,7 @@ int pblk_gc_sysfs_force(struct pblk *pblk, int force);
>> /*
>> * pblk rate limiter
>> */
>> -void pblk_rl_init(struct pblk_rl *rl, int budget);
>> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold);
>> void pblk_rl_free(struct pblk_rl *rl);
>> void pblk_rl_update_rates(struct pblk_rl *rl);
>> int pblk_rl_high_thrs(struct pblk_rl *rl);
>> --
>> 2.17.1
>>

2019-01-31 20:12:56

by Hans Holmberg

[permalink] [raw]
Subject: Re: [PATCH V2] lightnvm: pblk: prevent stall due to wb threshold

On Thu, Jan 31, 2019 at 5:33 PM Javier González <[email protected]> wrote:
>
>
>
> > On 31 Jan 2019, at 11.41, Hans Holmberg <[email protected]> wrote:
> >
> > Hi Javier!
> >
> > How did you test this? I'm trying to add a test case to our testing framework.
> >
> > This is what i ran in qemu, and I got a hang (with this version of the patch)
> >
> > nvme lnvm create -d nvme0n1 -t pblk -n pblk0 -f -b 0 -e 0
>
> I run several low configurations without problem. Can you share the qemu configuration and version?
>

Of course!

qemu remote: https://github.com/CNEX-Labs/qemu-nvme.git
branch: master (cb200e3ccf9c1ff21f6275c6cb68b2801135a640)

My geometry:
parallel units: 8, secs per chk: 768, meta size 16, ws min 12, ws opt
24, cunits=0

> I’m on travel until Friday - I’ll come back to you over the weekend.

Oh, no worries. This patch did not introduce the issue, so it's not a
regression, but it looks like the same type of hang as the patch
addresses.
I might have stumbled across another bug.

One thing I noticed was that the dd write block size that triggered
the hang is >= write buffer. (128 entries * 4k)

Safe travels!

>
> >
> > kernel log: [ 116.381799] pblk pblk0: luns:1, lines:280, secs:212736,
> > buf entries:128
> >
> > # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=4k count=1
> > 1+0 records in
> > 1+0 records out
> > 4096 bytes (4.1 kB, 4.0 KiB) copied, 0.000480941 s, 8.5 MB/s
> > # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=64k count=1
> > 1+0 records in
> > 1+0 records out
> > 65536 bytes (66 kB, 64 KiB) copied, 0.000477373 s, 137 MB/s
> > # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=128k count=1
> > 1+0 records in
> > 1+0 records out
> > 131072 bytes (131 kB, 128 KiB) copied, 0.000548722 s, 239 MB/s
> > # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=256k count=1
> > 1+0 records in
> > 1+0 records out
> > 262144 bytes (262 kB, 256 KiB) copied, 0.000718515 s, 365 MB/s
> > # dd if=/dev/zero of=/dev/pblk0 oflag=direct bs=512k count=1
> > <HANG>
>
>
> >
> >
> >> On Wed, Jan 30, 2019 at 11:28 AM Javier González <[email protected]> wrote:
> >>
> >> In order to respect mw_cuinits, pblk's write buffer maintains a
> >> backpointer to protect data not yet persisted; when writing to the write
> >> buffer, this backpointer defines a threshold that pblk's rate-limiter
> >> enforces.
> >>
> >> On small PU configurations, the following scenarios might take place: (i)
> >> the threshold is larger than the write buffer and (ii) the threshold is
> >> smaller than the write buffer, but larger than the maximun allowed
> >> split bio - 256KB at this moment (Note that writes are not always
> >> split - we only do this when we the size of the buffer is smaller
> >> than the buffer). In both cases, pblk's rate-limiter prevents the I/O to
> >> be written to the buffer, thus stalling.
> >>
> >> This patch fixes the original backpointer implementation by considering
> >> the threshold both on buffer creation and on the rate-limiters path,
> >> when bio_split is triggered (case (ii) above).
> >>
> >> Fixes: 766c8ceb16fc ("lightnvm: pblk: guarantee that backpointer is respected on writer stall")
> >> Signed-off-by: Javier González <[email protected]>
> >> ---
> >>
> >> Changes since V1:
> >> - Fix a bad arithmetinc on the rate-limiter max_io calculation (from
> >> Hans)
> >>
> >> drivers/lightnvm/pblk-rb.c | 25 +++++++++++++++++++------
> >> drivers/lightnvm/pblk-rl.c | 5 ++---
> >> drivers/lightnvm/pblk.h | 2 +-
> >> 3 files changed, 22 insertions(+), 10 deletions(-)
> >>
> >> diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c
> >> index d4ca8c64ee0f..a6133b50ed9c 100644
> >> --- a/drivers/lightnvm/pblk-rb.c
> >> +++ b/drivers/lightnvm/pblk-rb.c
> >> @@ -45,10 +45,23 @@ void pblk_rb_free(struct pblk_rb *rb)
> >> /*
> >> * pblk_rb_calculate_size -- calculate the size of the write buffer
> >> */
> >> -static unsigned int pblk_rb_calculate_size(unsigned int nr_entries)
> >> +static unsigned int pblk_rb_calculate_size(unsigned int nr_entries,
> >> + unsigned int threshold)
> >> {
> >> - /* Alloc a write buffer that can at least fit 128 entries */
> >> - return (1 << max(get_count_order(nr_entries), 7));
> >> + unsigned int thr_sz = 1 << (get_count_order(threshold + NVM_MAX_VLBA));
> >> + unsigned int max_sz = max(thr_sz, nr_entries);
> >> + unsigned int max_io;
> >> +
> >> + /* Alloc a write buffer that can (i) fit at least two split bios
> >> + * (considering max I/O size NVM_MAX_VLBA, and (ii) guarantee that the
> >> + * threshold will be respected
> >> + */
> >> + max_io = (1 << max((int)(get_count_order(max_sz)),
> >> + (int)(get_count_order(NVM_MAX_VLBA << 1))));
> >> + if ((threshold + NVM_MAX_VLBA) >= max_io)
> >> + max_io <<= 1;
> >> +
> >> + return max_io;
> >> }
> >>
> >> /*
> >> @@ -67,12 +80,12 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> >> unsigned int alloc_order, order, iter;
> >> unsigned int nr_entries;
> >>
> >> - nr_entries = pblk_rb_calculate_size(size);
> >> + nr_entries = pblk_rb_calculate_size(size, threshold);
> >> entries = vzalloc(array_size(nr_entries, sizeof(struct pblk_rb_entry)));
> >> if (!entries)
> >> return -ENOMEM;
> >>
> >> - power_size = get_count_order(size);
> >> + power_size = get_count_order(nr_entries);
> >> power_seg_sz = get_count_order(seg_size);
> >>
> >> down_write(&pblk_rb_lock);
> >> @@ -149,7 +162,7 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> >> * Initialize rate-limiter, which controls access to the write buffer
> >> * by user and GC I/O
> >> */
> >> - pblk_rl_init(&pblk->rl, rb->nr_entries);
> >> + pblk_rl_init(&pblk->rl, rb->nr_entries, threshold);
> >>
> >> return 0;
> >> }
> >> diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
> >> index 76116d5f78e4..e9e0af0df165 100644
> >> --- a/drivers/lightnvm/pblk-rl.c
> >> +++ b/drivers/lightnvm/pblk-rl.c
> >> @@ -207,7 +207,7 @@ void pblk_rl_free(struct pblk_rl *rl)
> >> del_timer(&rl->u_timer);
> >> }
> >>
> >> -void pblk_rl_init(struct pblk_rl *rl, int budget)
> >> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold)
> >> {
> >> struct pblk *pblk = container_of(rl, struct pblk, rl);
> >> struct nvm_tgt_dev *dev = pblk->dev;
> >> @@ -217,7 +217,6 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> >> int sec_meta, blk_meta;
> >> unsigned int rb_windows;
> >>
> >> -
> >> /* Consider sectors used for metadata */
> >> sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
> >> blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
> >> @@ -234,7 +233,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> >> /* To start with, all buffer is available to user I/O writers */
> >> rl->rb_budget = budget;
> >> rl->rb_user_max = budget;
> >> - rl->rb_max_io = budget >> 1;
> >> + rl->rb_max_io = budget - threshold;
> >> rl->rb_gc_max = 0;
> >> rl->rb_state = PBLK_RL_HIGH;
> >>
> >> diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
> >> index 72ae8755764e..a6386d5acd73 100644
> >> --- a/drivers/lightnvm/pblk.h
> >> +++ b/drivers/lightnvm/pblk.h
> >> @@ -924,7 +924,7 @@ int pblk_gc_sysfs_force(struct pblk *pblk, int force);
> >> /*
> >> * pblk rate limiter
> >> */
> >> -void pblk_rl_init(struct pblk_rl *rl, int budget);
> >> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold);
> >> void pblk_rl_free(struct pblk_rl *rl);
> >> void pblk_rl_update_rates(struct pblk_rl *rl);
> >> int pblk_rl_high_thrs(struct pblk_rl *rl);
> >> --
> >> 2.17.1
> >>

2019-02-04 08:16:19

by Javier González

[permalink] [raw]
Subject: Re: [PATCH V2] lightnvm: pblk: prevent stall due to wb threshold


> On 31 Jan 2019, at 21.10, Hans Holmberg <[email protected]> wrote:
>
> On Thu, Jan 31, 2019 at 5:33 PM Javier González <[email protected]> wrote:
>>> On 31 Jan 2019, at 11.41, Hans Holmberg <[email protected]> wrote:
>>>
>>> Hi Javier!
>>>
>>> How did you test this? I'm trying to add a test case to our testing framework.
>>>
>>> This is what i ran in qemu, and I got a hang (with this version of the patch)
>>>
>>> nvme lnvm create -d nvme0n1 -t pblk -n pblk0 -f -b 0 -e 0
>>
>> I run several low configurations without problem. Can you share the qemu configuration and version?
>
> Of course!
>
> qemu remote: https://github.com/CNEX-Labs/qemu-nvme.git
> branch: master (cb200e3ccf9c1ff21f6275c6cb68b2801135a640)
>
> My geometry:
> parallel units: 8, secs per chk: 768, meta size 16, ws min 12, ws opt
> 24, cunits=0

You are right, lmw_cunits=0 gives the problem.

What about the following? If it goes through cijoe, I'll send a V3.

diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c
index d4ca8c64ee0f..a6133b50ed9c 100644
--- a/drivers/lightnvm/pblk-rb.c
+++ b/drivers/lightnvm/pblk-rb.c
@@ -45,10 +45,23 @@ void pblk_rb_free(struct pblk_rb *rb)
/*
* pblk_rb_calculate_size -- calculate the size of the write buffer
*/
-static unsigned int pblk_rb_calculate_size(unsigned int nr_entries)
+static unsigned int pblk_rb_calculate_size(unsigned int nr_entries,
+ unsigned int threshold)
{
- /* Alloc a write buffer that can at least fit 128 entries */
- return (1 << max(get_count_order(nr_entries), 7));
+ unsigned int thr_sz = 1 << (get_count_order(threshold + NVM_MAX_VLBA));
+ unsigned int max_sz = max(thr_sz, nr_entries);
+ unsigned int max_io;
+
+ /* Alloc a write buffer that can (i) fit at least two split bios
+ * (considering max I/O size NVM_MAX_VLBA, and (ii) guarantee that the
+ * threshold will be respected
+ */
+ max_io = (1 << max((int)(get_count_order(max_sz)),
+ (int)(get_count_order(NVM_MAX_VLBA << 1))));
+ if ((threshold + NVM_MAX_VLBA) >= max_io)
+ max_io <<= 1;
+
+ return max_io;
}

/*
@@ -67,12 +80,12 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
unsigned int alloc_order, order, iter;
unsigned int nr_entries;

- nr_entries = pblk_rb_calculate_size(size);
+ nr_entries = pblk_rb_calculate_size(size, threshold);
entries = vzalloc(array_size(nr_entries, sizeof(struct pblk_rb_entry)));
if (!entries)
return -ENOMEM;

- power_size = get_count_order(size);
+ power_size = get_count_order(nr_entries);
power_seg_sz = get_count_order(seg_size);

down_write(&pblk_rb_lock);
@@ -149,7 +162,7 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
* Initialize rate-limiter, which controls access to the write buffer
* by user and GC I/O
*/
- pblk_rl_init(&pblk->rl, rb->nr_entries);
+ pblk_rl_init(&pblk->rl, rb->nr_entries, threshold);

return 0;
}
diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
index 76116d5f78e4..b014957dde0b 100644
--- a/drivers/lightnvm/pblk-rl.c
+++ b/drivers/lightnvm/pblk-rl.c
@@ -207,7 +207,7 @@ void pblk_rl_free(struct pblk_rl *rl)
del_timer(&rl->u_timer);
}

-void pblk_rl_init(struct pblk_rl *rl, int budget)
+void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold)
{
struct pblk *pblk = container_of(rl, struct pblk, rl);
struct nvm_tgt_dev *dev = pblk->dev;
@@ -217,7 +217,6 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
int sec_meta, blk_meta;
unsigned int rb_windows;

-
/* Consider sectors used for metadata */
sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
@@ -234,7 +233,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
/* To start with, all buffer is available to user I/O writers */
rl->rb_budget = budget;
rl->rb_user_max = budget;
- rl->rb_max_io = budget >> 1;
+ rl->rb_max_io = threshold ? (budget - threshold) : (budget - 1);
rl->rb_gc_max = 0;
rl->rb_state = PBLK_RL_HIGH;

diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
index 72ae8755764e..a6386d5acd73 100644
--- a/drivers/lightnvm/pblk.h
+++ b/drivers/lightnvm/pblk.h
@@ -924,7 +924,7 @@ int pblk_gc_sysfs_force(struct pblk *pblk, int force);
/*
* pblk rate limiter
*/
-void pblk_rl_init(struct pblk_rl *rl, int budget);
+void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold);
void pblk_rl_free(struct pblk_rl *rl);
void pblk_rl_update_rates(struct pblk_rl *rl);
int pblk_rl_high_thrs(struct pblk_rl *rl);
--
2.17.1


Attachments:
signature.asc (849.00 B)
Message signed with OpenPGP

2019-02-04 21:51:45

by Hans Holmberg

[permalink] [raw]
Subject: Re: [PATCH V2] lightnvm: pblk: prevent stall due to wb threshold

On Mon, Feb 4, 2019 at 9:14 AM Javier González <[email protected]> wrote:
>
>
> > On 31 Jan 2019, at 21.10, Hans Holmberg <[email protected]> wrote:
> >
> > On Thu, Jan 31, 2019 at 5:33 PM Javier González <[email protected]> wrote:
> >>> On 31 Jan 2019, at 11.41, Hans Holmberg <[email protected]> wrote:
> >>>
> >>> Hi Javier!
> >>>
> >>> How did you test this? I'm trying to add a test case to our testing framework.
> >>>
> >>> This is what i ran in qemu, and I got a hang (with this version of the patch)
> >>>
> >>> nvme lnvm create -d nvme0n1 -t pblk -n pblk0 -f -b 0 -e 0
> >>
> >> I run several low configurations without problem. Can you share the qemu configuration and version?
> >
> > Of course!
> >
> > qemu remote: https://github.com/CNEX-Labs/qemu-nvme.git
> > branch: master (cb200e3ccf9c1ff21f6275c6cb68b2801135a640)
> >
> > My geometry:
> > parallel units: 8, secs per chk: 768, meta size 16, ws min 12, ws opt
> > 24, cunits=0
>
> You are right, lmw_cunits=0 gives the problem.
>
> What about the following? If it goes through cijoe, I'll send a V3.

Please do send a V3 if it passes your testing.

Thanks!
Hans

>
> diff --git a/drivers/lightnvm/pblk-rb.c b/drivers/lightnvm/pblk-rb.c
> index d4ca8c64ee0f..a6133b50ed9c 100644
> --- a/drivers/lightnvm/pblk-rb.c
> +++ b/drivers/lightnvm/pblk-rb.c
> @@ -45,10 +45,23 @@ void pblk_rb_free(struct pblk_rb *rb)
> /*
> * pblk_rb_calculate_size -- calculate the size of the write buffer
> */
> -static unsigned int pblk_rb_calculate_size(unsigned int nr_entries)
> +static unsigned int pblk_rb_calculate_size(unsigned int nr_entries,
> + unsigned int threshold)
> {
> - /* Alloc a write buffer that can at least fit 128 entries */
> - return (1 << max(get_count_order(nr_entries), 7));
> + unsigned int thr_sz = 1 << (get_count_order(threshold + NVM_MAX_VLBA));
> + unsigned int max_sz = max(thr_sz, nr_entries);
> + unsigned int max_io;
> +
> + /* Alloc a write buffer that can (i) fit at least two split bios
> + * (considering max I/O size NVM_MAX_VLBA, and (ii) guarantee that the
> + * threshold will be respected
> + */
> + max_io = (1 << max((int)(get_count_order(max_sz)),
> + (int)(get_count_order(NVM_MAX_VLBA << 1))));
> + if ((threshold + NVM_MAX_VLBA) >= max_io)
> + max_io <<= 1;
> +
> + return max_io;
> }
>
> /*
> @@ -67,12 +80,12 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> unsigned int alloc_order, order, iter;
> unsigned int nr_entries;
>
> - nr_entries = pblk_rb_calculate_size(size);
> + nr_entries = pblk_rb_calculate_size(size, threshold);
> entries = vzalloc(array_size(nr_entries, sizeof(struct pblk_rb_entry)));
> if (!entries)
> return -ENOMEM;
>
> - power_size = get_count_order(size);
> + power_size = get_count_order(nr_entries);
> power_seg_sz = get_count_order(seg_size);
>
> down_write(&pblk_rb_lock);
> @@ -149,7 +162,7 @@ int pblk_rb_init(struct pblk_rb *rb, unsigned int size, unsigned int threshold,
> * Initialize rate-limiter, which controls access to the write buffer
> * by user and GC I/O
> */
> - pblk_rl_init(&pblk->rl, rb->nr_entries);
> + pblk_rl_init(&pblk->rl, rb->nr_entries, threshold);
>
> return 0;
> }
> diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
> index 76116d5f78e4..b014957dde0b 100644
> --- a/drivers/lightnvm/pblk-rl.c
> +++ b/drivers/lightnvm/pblk-rl.c
> @@ -207,7 +207,7 @@ void pblk_rl_free(struct pblk_rl *rl)
> del_timer(&rl->u_timer);
> }
>
> -void pblk_rl_init(struct pblk_rl *rl, int budget)
> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold)
> {
> struct pblk *pblk = container_of(rl, struct pblk, rl);
> struct nvm_tgt_dev *dev = pblk->dev;
> @@ -217,7 +217,6 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> int sec_meta, blk_meta;
> unsigned int rb_windows;
>
> -
> /* Consider sectors used for metadata */
> sec_meta = (lm->smeta_sec + lm->emeta_sec[0]) * l_mg->nr_free_lines;
> blk_meta = DIV_ROUND_UP(sec_meta, geo->clba);
> @@ -234,7 +233,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
> /* To start with, all buffer is available to user I/O writers */
> rl->rb_budget = budget;
> rl->rb_user_max = budget;
> - rl->rb_max_io = budget >> 1;
> + rl->rb_max_io = threshold ? (budget - threshold) : (budget - 1);
> rl->rb_gc_max = 0;
> rl->rb_state = PBLK_RL_HIGH;
>
> diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
> index 72ae8755764e..a6386d5acd73 100644
> --- a/drivers/lightnvm/pblk.h
> +++ b/drivers/lightnvm/pblk.h
> @@ -924,7 +924,7 @@ int pblk_gc_sysfs_force(struct pblk *pblk, int force);
> /*
> * pblk rate limiter
> */
> -void pblk_rl_init(struct pblk_rl *rl, int budget);
> +void pblk_rl_init(struct pblk_rl *rl, int budget, int threshold);
> void pblk_rl_free(struct pblk_rl *rl);
> void pblk_rl_update_rates(struct pblk_rl *rl);
> int pblk_rl_high_thrs(struct pblk_rl *rl);
> --
> 2.17.1
>