Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Javier Gonzalez <javier@cnexlabs.com>
To:     Javier Gonzalez <javier@cnexlabs.com>
CC:     Hans Holmberg <hans.ml.holmberg@owltronix.com>,
        =?utf-8?B?TWF0aWFzIEJqw7hybGluZw==?= <mb@lightnvm.io>,
        "linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Hans Holmberg <hans.holmberg@cnexlabs.com>
Subject: Re: [PATCH v2 2/3] lightnvm: pblk: garbage collect lines with failed
 writes
Thread-Topic: [PATCH v2 2/3] lightnvm: pblk: garbage collect lines with failed
 writes
Thread-Index: AQHT24+ChfawhRlNy0+cu7vWPMNoyKQZEDOAgAABS4A=
Date:   Mon, 30 Apr 2018 09:19:19 +0000
Message-ID: <59DA13E5-2CD0-4C95-BE11-C53A4755BF79@cnexlabs.com>
References: <1524548732-4326-1-git-send-email-hans.ml.holmberg@owltronix.com>
 <1524548732-4326-3-git-send-email-hans.ml.holmberg@owltronix.com>
 <269EA9FA-DD9B-4D22-A7F6-6AB8B937D930@cnexlabs.com>
In-Reply-To: <269EA9FA-DD9B-4D22-A7F6-6AB8B937D930@cnexlabs.com>
Accept-Language: en-US
Content-Language: en-US
received-spf: None (protection.outlook.com: cnexlabs.com does not designate
 permitted sender hosts)
spamdiagnosticoutput: 1:99
spamdiagnosticmetadata: NSPM
Content-Type: multipart/signed;
        boundary="Apple-Mail=_033826BA-F62C-4A3B-8E14-C277C2A51A44";
        protocol="application/pgp-signature"; micalg=pgp-sha512
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 1e09434a-1caf-4c27-ec8a-08d5ae7b7485
X-MS-Exchange-CrossTenant-originalarrivaltime: 30 Apr 2018 09:19:19.9281
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: e40dfc2e-c6c1-463a-a598-38602b2c3cff
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CO2PR06MB489
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

--Apple-Mail=_033826BA-F62C-4A3B-8E14-C277C2A51A44
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
	charset=utf-8


> On 30 Apr 2018, at 11.14, Javier Gonzalez <javier@cnexlabs.com> wrote:
>=20
>> On 24 Apr 2018, at 07.45, Hans Holmberg =
<hans.ml.holmberg@owltronix.com> wrote:
>>=20
>> From: Hans Holmberg <hans.holmberg@cnexlabs.com>
>>=20
>> Write failures should not happen under normal circumstances,
>> so in order to bring the chunk back into a known state as soon
>> as possible, evacuate all the valid data out of the line and let the
>> fw judge if the block can be written to in the next reset cycle.
>>=20
>> Do this by introducing a new gc list for lines with failed writes,
>> and ensure that the rate limiter allocates a small portion of
>> the write bandwidth to get the job done.
>>=20
>> The lba list is saved in memory for use during gc as we
>> cannot gurantee that the emeta data is readable if a write
>> error occurred.
>>=20
>> Signed-off-by: Hans Holmberg <hans.holmberg@cnexlabs.com>
>> ---
>> drivers/lightnvm/pblk-core.c  |  45 ++++++++++++++++++-
>> drivers/lightnvm/pblk-gc.c    | 102 =
+++++++++++++++++++++++++++---------------
>> drivers/lightnvm/pblk-init.c  |  45 ++++++++++++-------
>> drivers/lightnvm/pblk-rl.c    |  29 ++++++++++--
>> drivers/lightnvm/pblk-sysfs.c |  15 ++++++-
>> drivers/lightnvm/pblk-write.c |   2 +
>> drivers/lightnvm/pblk.h       |  25 +++++++++--
>> 7 files changed, 199 insertions(+), 64 deletions(-)
>>=20
>> diff --git a/drivers/lightnvm/pblk-core.c =
b/drivers/lightnvm/pblk-core.c
>> index 7762e89..413cf3b 100644
>> --- a/drivers/lightnvm/pblk-core.c
>> +++ b/drivers/lightnvm/pblk-core.c
>> @@ -373,7 +373,13 @@ struct list_head *pblk_line_gc_list(struct pblk =
*pblk, struct pblk_line *line)
>>=20
>> 	lockdep_assert_held(&line->lock);
>>=20
>> -	if (!vsc) {
>> +	if (line->w_err_gc->has_write_err) {
>> +		if (line->gc_group !=3D PBLK_LINEGC_WERR) {
>> +			line->gc_group =3D PBLK_LINEGC_WERR;
>> +			move_list =3D &l_mg->gc_werr_list;
>> +			pblk_rl_werr_line_in(&pblk->rl);
>> +		}
>> +	} else if (!vsc) {
>> 		if (line->gc_group !=3D PBLK_LINEGC_FULL) {
>> 			line->gc_group =3D PBLK_LINEGC_FULL;
>> 			move_list =3D &l_mg->gc_full_list;
>> @@ -1603,8 +1609,13 @@ static void __pblk_line_put(struct pblk *pblk, =
struct pblk_line *line)
>> 	line->state =3D PBLK_LINESTATE_FREE;
>> 	line->gc_group =3D PBLK_LINEGC_NONE;
>> 	pblk_line_free(line);
>> -	spin_unlock(&line->lock);
>>=20
>> +	if (line->w_err_gc->has_write_err) {
>> +		pblk_rl_werr_line_out(&pblk->rl);
>> +		line->w_err_gc->has_write_err =3D 0;
>> +	}
>> +
>> +	spin_unlock(&line->lock);
>> 	atomic_dec(&gc->pipeline_gc);
>>=20
>> 	spin_lock(&l_mg->free_lock);
>> @@ -1767,11 +1778,34 @@ void pblk_line_close_meta(struct pblk *pblk, =
struct pblk_line *line)
>>=20
>> 	spin_lock(&l_mg->close_lock);
>> 	spin_lock(&line->lock);
>> +
>> +	/* Update the in-memory start address for emeta, in case it has
>> +	 * shifted due to write errors
>> +	 */
>> +	if (line->emeta_ssec !=3D line->cur_sec)
>> +		line->emeta_ssec =3D line->cur_sec;
>> +
>> 	list_add_tail(&line->list, &l_mg->emeta_list);
>> 	spin_unlock(&line->lock);
>> 	spin_unlock(&l_mg->close_lock);
>>=20
>> 	pblk_line_should_sync_meta(pblk);
>> +
>> +
>> +}
>> +
>> +static void pblk_save_lba_list(struct pblk *pblk, struct pblk_line =
*line)
>> +{
>> +	struct pblk_line_meta *lm =3D &pblk->lm;
>> +	struct pblk_line_mgmt *l_mg =3D &pblk->l_mg;
>> +	unsigned int lba_list_size =3D lm->emeta_len[2];
>> +	struct pblk_w_err_gc *w_err_gc =3D line->w_err_gc;
>> +	struct pblk_emeta *emeta =3D line->emeta;
>> +
>> +	w_err_gc->lba_list =3D pblk_malloc(lba_list_size,
>> +					 l_mg->emeta_alloc_type, =
GFP_KERNEL);
>> +	memcpy(w_err_gc->lba_list, emeta_to_lbas(pblk, emeta->buf),
>> +				lba_list_size);
>> }
>>=20
>> void pblk_line_close_ws(struct work_struct *work)
>> @@ -1780,6 +1814,13 @@ void pblk_line_close_ws(struct work_struct =
*work)
>> 									=
ws);
>> 	struct pblk *pblk =3D line_ws->pblk;
>> 	struct pblk_line *line =3D line_ws->line;
>> +	struct pblk_w_err_gc *w_err_gc =3D line->w_err_gc;
>> +
>> +	/* Write errors makes the emeta start address stored in smeta =
invalid,
>> +	 * so keep a copy of the lba list until we've gc'd the line
>> +	 */
>> +	if (w_err_gc->has_write_err)
>> +		pblk_save_lba_list(pblk, line);
>>=20
>> 	pblk_line_close(pblk, line);
>> 	mempool_free(line_ws, pblk->gen_ws_pool);
>> diff --git a/drivers/lightnvm/pblk-gc.c b/drivers/lightnvm/pblk-gc.c
>> index b0cc277..df88f1b 100644
>> --- a/drivers/lightnvm/pblk-gc.c
>> +++ b/drivers/lightnvm/pblk-gc.c
>> @@ -129,6 +129,53 @@ static void pblk_gc_line_ws(struct work_struct =
*work)
>> 	kfree(gc_rq_ws);
>> }
>>=20
>> +static __le64 *get_lba_list_from_emeta(struct pblk *pblk,
>> +				       struct pblk_line *line)
>> +{
>> +	struct line_emeta *emeta_buf;
>> +	struct pblk_line_mgmt *l_mg =3D &pblk->l_mg;
>> +	struct pblk_line_meta *lm =3D &pblk->lm;
>> +	unsigned int lba_list_size =3D lm->emeta_len[2];
>> +	__le64 *lba_list;
>> +	int ret;
>> +
>> +	emeta_buf =3D pblk_malloc(lm->emeta_len[0],
>> +				l_mg->emeta_alloc_type, GFP_KERNEL);
>> +	if (!emeta_buf)
>> +		return NULL;
>> +
>> +	ret =3D pblk_line_read_emeta(pblk, line, emeta_buf);
>> +	if (ret) {
>> +		pr_err("pblk: line %d read emeta failed (%d)\n",
>> +				line->id, ret);
>> +		pblk_mfree(emeta_buf, l_mg->emeta_alloc_type);
>> +		return NULL;
>> +	}
>> +
>> +	/* If this read fails, it means that emeta is corrupted.
>> +	 * For now, leave the line untouched.
>> +	 * TODO: Implement a recovery routine that scans and moves
>> +	 * all sectors on the line.
>> +	 */
>> +
>> +	ret =3D pblk_recov_check_emeta(pblk, emeta_buf);
>> +	if (ret) {
>> +		pr_err("pblk: inconsistent emeta (line %d)\n",
>> +				line->id);
>> +		pblk_mfree(emeta_buf, l_mg->emeta_alloc_type);
>> +		return NULL;
>> +	}
>> +
>> +	lba_list =3D pblk_malloc(lba_list_size,
>> +			       l_mg->emeta_alloc_type, GFP_KERNEL);
>> +	if (lba_list)
>> +		memcpy(lba_list, emeta_to_lbas(pblk, emeta_buf), =
lba_list_size);
>> +
>> +	pblk_mfree(emeta_buf, l_mg->emeta_alloc_type);
>> +
>> +	return lba_list;
>> +}
>> +
>> static void pblk_gc_line_prepare_ws(struct work_struct *work)
>> {
>> 	struct pblk_line_ws *line_ws =3D container_of(work, struct =
pblk_line_ws,
>> @@ -138,46 +185,26 @@ static void pblk_gc_line_prepare_ws(struct =
work_struct *work)
>> 	struct pblk_line_mgmt *l_mg =3D &pblk->l_mg;
>> 	struct pblk_line_meta *lm =3D &pblk->lm;
>> 	struct pblk_gc *gc =3D &pblk->gc;
>> -	struct line_emeta *emeta_buf;
>> 	struct pblk_line_ws *gc_rq_ws;
>> 	struct pblk_gc_rq *gc_rq;
>> 	__le64 *lba_list;
>> 	unsigned long *invalid_bitmap;
>> 	int sec_left, nr_secs, bit;
>> -	int ret;
>>=20
>> 	invalid_bitmap =3D kmalloc(lm->sec_bitmap_len, GFP_KERNEL);
>> 	if (!invalid_bitmap)
>> 		goto fail_free_ws;
>>=20
>> -	emeta_buf =3D pblk_malloc(lm->emeta_len[0], =
l_mg->emeta_alloc_type,
>> -								=
GFP_KERNEL);
>> -	if (!emeta_buf) {
>> -		pr_err("pblk: cannot use GC emeta\n");
>> -		goto fail_free_bitmap;
>> -	}
>> -
>> -	ret =3D pblk_line_read_emeta(pblk, line, emeta_buf);
>> -	if (ret) {
>> -		pr_err("pblk: line %d read emeta failed (%d)\n", =
line->id, ret);
>> -		goto fail_free_emeta;
>> -	}
>> -
>> -	/* If this read fails, it means that emeta is corrupted. For =
now, leave
>> -	 * the line untouched. TODO: Implement a recovery routine that =
scans and
>> -	 * moves all sectors on the line.
>> -	 */
>> -
>> -	ret =3D pblk_recov_check_emeta(pblk, emeta_buf);
>> -	if (ret) {
>> -		pr_err("pblk: inconsistent emeta (line %d)\n", =
line->id);
>> -		goto fail_free_emeta;
>> -	}
>> -
>> -	lba_list =3D emeta_to_lbas(pblk, emeta_buf);
>> -	if (!lba_list) {
>> -		pr_err("pblk: could not interpret emeta (line %d)\n", =
line->id);
>> -		goto fail_free_emeta;
>> +	if (line->w_err_gc->has_write_err) {
>> +		lba_list =3D line->w_err_gc->lba_list;
>> +		line->w_err_gc->lba_list =3D NULL;
>> +	} else {
>> +		lba_list =3D get_lba_list_from_emeta(pblk, line);
>> +		if (!lba_list) {
>> +			pr_err("pblk: could not interpret emeta (line =
%d)\n",
>> +					line->id);
>> +			goto fail_free_ws;
>> +		}
>> 	}
>>=20
>> 	spin_lock(&line->lock);
>> @@ -187,14 +214,14 @@ static void pblk_gc_line_prepare_ws(struct =
work_struct *work)
>>=20
>> 	if (sec_left < 0) {
>> 		pr_err("pblk: corrupted GC line (%d)\n", line->id);
>> -		goto fail_free_emeta;
>> +		goto fail_free_lba_list;
>> 	}
>>=20
>> 	bit =3D -1;
>> next_rq:
>> 	gc_rq =3D kmalloc(sizeof(struct pblk_gc_rq), GFP_KERNEL);
>> 	if (!gc_rq)
>> -		goto fail_free_emeta;
>> +		goto fail_free_lba_list;
>>=20
>> 	nr_secs =3D 0;
>> 	do {
>> @@ -240,7 +267,7 @@ static void pblk_gc_line_prepare_ws(struct =
work_struct *work)
>> 		goto next_rq;
>>=20
>> out:
>> -	pblk_mfree(emeta_buf, l_mg->emeta_alloc_type);
>> +	pblk_mfree(lba_list, l_mg->emeta_alloc_type);
>> 	kfree(line_ws);
>> 	kfree(invalid_bitmap);
>>=20
>> @@ -251,9 +278,8 @@ static void pblk_gc_line_prepare_ws(struct =
work_struct *work)
>>=20
>> fail_free_gc_rq:
>> 	kfree(gc_rq);
>> -fail_free_emeta:
>> -	pblk_mfree(emeta_buf, l_mg->emeta_alloc_type);
>> -fail_free_bitmap:
>> +fail_free_lba_list:
>> +	pblk_mfree(lba_list, l_mg->emeta_alloc_type);
>> 	kfree(invalid_bitmap);
>> fail_free_ws:
>> 	kfree(line_ws);
>> @@ -349,12 +375,14 @@ static struct pblk_line =
*pblk_gc_get_victim_line(struct pblk *pblk,
>> static bool pblk_gc_should_run(struct pblk_gc *gc, struct pblk_rl =
*rl)
>> {
>> 	unsigned int nr_blocks_free, nr_blocks_need;
>> +	unsigned int werr_lines =3D atomic_read(&rl->werr_lines);
>>=20
>> 	nr_blocks_need =3D pblk_rl_high_thrs(rl);
>> 	nr_blocks_free =3D pblk_rl_nr_free_blks(rl);
>>=20
>> 	/* This is not critical, no need to take lock here */
>> -	return ((gc->gc_active) && (nr_blocks_need > nr_blocks_free));
>> +	return ((werr_lines > 0) ||
>> +		((gc->gc_active) && (nr_blocks_need > nr_blocks_free)));
>> }
>>=20
>> void pblk_gc_free_full_lines(struct pblk *pblk)
>> diff --git a/drivers/lightnvm/pblk-init.c =
b/drivers/lightnvm/pblk-init.c
>> index 6f06727..931ba32 100644
>> --- a/drivers/lightnvm/pblk-init.c
>> +++ b/drivers/lightnvm/pblk-init.c
>> @@ -493,11 +493,16 @@ static void pblk_line_mg_free(struct pblk =
*pblk)
>> 	}
>> }
>>=20
>> -static void pblk_line_meta_free(struct pblk_line *line)
>> +static void pblk_line_meta_free(struct pblk_line_mgmt *l_mg, struct =
pblk_line *line)

Actually, this goes over 80 lines - please run checkpatch.

Matias: can you fix this when picking it up? Thanks!


>> {
>> +	struct pblk_w_err_gc *w_err_gc =3D line->w_err_gc;
>> +
>> 	kfree(line->blk_bitmap);
>> 	kfree(line->erase_bitmap);
>> 	kfree(line->chks);
>> +
>> +	pblk_mfree(w_err_gc->lba_list, l_mg->emeta_alloc_type);
>> +	kfree(w_err_gc);
>> }
>>=20
>> static void pblk_lines_free(struct pblk *pblk)
>> @@ -511,7 +516,7 @@ static void pblk_lines_free(struct pblk *pblk)
>> 		line =3D &pblk->lines[i];
>>=20
>> 		pblk_line_free(line);
>> -		pblk_line_meta_free(line);
>> +		pblk_line_meta_free(l_mg, line);
>> 	}
>> 	spin_unlock(&l_mg->free_lock);
>>=20
>> @@ -813,20 +818,28 @@ static int pblk_alloc_line_meta(struct pblk =
*pblk, struct pblk_line *line)
>> 		return -ENOMEM;
>>=20
>> 	line->erase_bitmap =3D kzalloc(lm->blk_bitmap_len, GFP_KERNEL);
>> -	if (!line->erase_bitmap) {
>> -		kfree(line->blk_bitmap);
>> -		return -ENOMEM;
>> -	}
>> +	if (!line->erase_bitmap)
>> +		goto free_blk_bitmap;
>> +
>>=20
>> 	line->chks =3D kmalloc(lm->blk_per_line * sizeof(struct =
nvm_chk_meta),
>> 								=
GFP_KERNEL);
>> -	if (!line->chks) {
>> -		kfree(line->erase_bitmap);
>> -		kfree(line->blk_bitmap);
>> -		return -ENOMEM;
>> -	}
>> +	if (!line->chks)
>> +		goto free_erase_bitmap;
>> +
>> +	line->w_err_gc =3D kzalloc(sizeof(struct pblk_w_err_gc), =
GFP_KERNEL);
>> +	if (!line->w_err_gc)
>> +		goto free_chks;
>>=20
>> 	return 0;
>> +
>> +free_chks:
>> +	kfree(line->chks);
>> +free_erase_bitmap:
>> +	kfree(line->erase_bitmap);
>> +free_blk_bitmap:
>> +	kfree(line->blk_bitmap);
>> +	return -ENOMEM;
>> }
>>=20
>> static int pblk_line_mg_init(struct pblk *pblk)
>> @@ -851,12 +864,14 @@ static int pblk_line_mg_init(struct pblk *pblk)
>> 	INIT_LIST_HEAD(&l_mg->gc_mid_list);
>> 	INIT_LIST_HEAD(&l_mg->gc_low_list);
>> 	INIT_LIST_HEAD(&l_mg->gc_empty_list);
>> +	INIT_LIST_HEAD(&l_mg->gc_werr_list);
>>=20
>> 	INIT_LIST_HEAD(&l_mg->emeta_list);
>>=20
>> -	l_mg->gc_lists[0] =3D &l_mg->gc_high_list;
>> -	l_mg->gc_lists[1] =3D &l_mg->gc_mid_list;
>> -	l_mg->gc_lists[2] =3D &l_mg->gc_low_list;
>> +	l_mg->gc_lists[0] =3D &l_mg->gc_werr_list;
>> +	l_mg->gc_lists[1] =3D &l_mg->gc_high_list;
>> +	l_mg->gc_lists[2] =3D &l_mg->gc_mid_list;
>> +	l_mg->gc_lists[3] =3D &l_mg->gc_low_list;
>>=20
>> 	spin_lock_init(&l_mg->free_lock);
>> 	spin_lock_init(&l_mg->close_lock);
>> @@ -1063,7 +1078,7 @@ static int pblk_lines_init(struct pblk *pblk)
>>=20
>> fail_free_lines:
>> 	while (--i >=3D 0)
>> -		pblk_line_meta_free(&pblk->lines[i]);
>> +		pblk_line_meta_free(l_mg, &pblk->lines[i]);
>> 	kfree(pblk->lines);
>> fail_free_chunk_meta:
>> 	kfree(chunk_meta);
>> diff --git a/drivers/lightnvm/pblk-rl.c b/drivers/lightnvm/pblk-rl.c
>> index 883a711..6a0616a 100644
>> --- a/drivers/lightnvm/pblk-rl.c
>> +++ b/drivers/lightnvm/pblk-rl.c
>> @@ -73,6 +73,16 @@ void pblk_rl_user_in(struct pblk_rl *rl, int =
nr_entries)
>> 	pblk_rl_kick_u_timer(rl);
>> }
>>=20
>> +void pblk_rl_werr_line_in(struct pblk_rl *rl)
>> +{
>> +	atomic_inc(&rl->werr_lines);
>> +}
>> +
>> +void pblk_rl_werr_line_out(struct pblk_rl *rl)
>> +{
>> +	atomic_dec(&rl->werr_lines);
>> +}
>> +
>> void pblk_rl_gc_in(struct pblk_rl *rl, int nr_entries)
>> {
>> 	atomic_add(nr_entries, &rl->rb_gc_cnt);
>> @@ -99,11 +109,21 @@ static void __pblk_rl_update_rates(struct =
pblk_rl *rl,
>> {
>> 	struct pblk *pblk =3D container_of(rl, struct pblk, rl);
>> 	int max =3D rl->rb_budget;
>> +	int werr_gc_needed =3D atomic_read(&rl->werr_lines);
>>=20
>> 	if (free_blocks >=3D rl->high) {
>> -		rl->rb_user_max =3D max;
>> -		rl->rb_gc_max =3D 0;
>> -		rl->rb_state =3D PBLK_RL_HIGH;
>> +		if (werr_gc_needed) {
>> +			/* Allocate a small budget for recovering
>> +			 * lines with write errors
>> +			 */
>> +			rl->rb_gc_max =3D 1 << rl->rb_windows_pw;
>> +			rl->rb_user_max =3D max - rl->rb_gc_max;
>> +			rl->rb_state =3D PBLK_RL_WERR;
>> +		} else {
>> +			rl->rb_user_max =3D max;
>> +			rl->rb_gc_max =3D 0;
>> +			rl->rb_state =3D PBLK_RL_OFF;
>> +		}
>> 	} else if (free_blocks < rl->high) {
>> 		int shift =3D rl->high_pw - rl->rb_windows_pw;
>> 		int user_windows =3D free_blocks >> shift;
>> @@ -124,7 +144,7 @@ static void __pblk_rl_update_rates(struct pblk_rl =
*rl,
>> 		rl->rb_state =3D PBLK_RL_LOW;
>> 	}
>>=20
>> -	if (rl->rb_state =3D=3D (PBLK_RL_MID | PBLK_RL_LOW))
>> +	if (rl->rb_state !=3D PBLK_RL_OFF)
>> 		pblk_gc_should_start(pblk);
>> 	else
>> 		pblk_gc_should_stop(pblk);
>> @@ -221,6 +241,7 @@ void pblk_rl_init(struct pblk_rl *rl, int budget)
>> 	atomic_set(&rl->rb_user_cnt, 0);
>> 	atomic_set(&rl->rb_gc_cnt, 0);
>> 	atomic_set(&rl->rb_space, -1);
>> +	atomic_set(&rl->werr_lines, 0);
>>=20
>> 	timer_setup(&rl->u_timer, pblk_rl_u_timer, 0);
>>=20
>> diff --git a/drivers/lightnvm/pblk-sysfs.c =
b/drivers/lightnvm/pblk-sysfs.c
>> index e61909a..88a0a7c 100644
>> --- a/drivers/lightnvm/pblk-sysfs.c
>> +++ b/drivers/lightnvm/pblk-sysfs.c
>> @@ -173,6 +173,8 @@ static ssize_t pblk_sysfs_lines(struct pblk =
*pblk, char *page)
>> 	int free_line_cnt =3D 0, closed_line_cnt =3D 0, emeta_line_cnt =3D=
 0;
>> 	int d_line_cnt =3D 0, l_line_cnt =3D 0;
>> 	int gc_full =3D 0, gc_high =3D 0, gc_mid =3D 0, gc_low =3D 0, =
gc_empty =3D 0;
>> +	int gc_werr =3D 0;
>> +
>> 	int bad =3D 0, cor =3D 0;
>> 	int msecs =3D 0, cur_sec =3D 0, vsc =3D 0, sec_in_line =3D 0;
>> 	int map_weight =3D 0, meta_weight =3D 0;
>> @@ -237,6 +239,15 @@ static ssize_t pblk_sysfs_lines(struct pblk =
*pblk, char *page)
>> 		gc_empty++;
>> 	}
>>=20
>> +	list_for_each_entry(line, &l_mg->gc_werr_list, list) {
>> +		if (line->type =3D=3D PBLK_LINETYPE_DATA)
>> +			d_line_cnt++;
>> +		else if (line->type =3D=3D PBLK_LINETYPE_LOG)
>> +			l_line_cnt++;
>> +		closed_line_cnt++;
>> +		gc_werr++;
>> +	}
>> +
>> 	list_for_each_entry(line, &l_mg->bad_list, list)
>> 		bad++;
>> 	list_for_each_entry(line, &l_mg->corrupt_list, list)
>> @@ -275,8 +286,8 @@ static ssize_t pblk_sysfs_lines(struct pblk =
*pblk, char *page)
>> 					l_mg->nr_lines);
>>=20
>> 	sz +=3D snprintf(page + sz, PAGE_SIZE - sz,
>> -		"GC: full:%d, high:%d, mid:%d, low:%d, empty:%d, =
queue:%d\n",
>> -			gc_full, gc_high, gc_mid, gc_low, gc_empty,
>> +		"GC: full:%d, high:%d, mid:%d, low:%d, empty:%d, werr: =
%d, queue:%d\n",
>> +			gc_full, gc_high, gc_mid, gc_low, gc_empty, =
gc_werr,
>> 			atomic_read(&pblk->gc.read_inflight_gc));
>>=20
>> 	sz +=3D snprintf(page + sz, PAGE_SIZE - sz,
>> diff --git a/drivers/lightnvm/pblk-write.c =
b/drivers/lightnvm/pblk-write.c
>> index f62e432f..f33c2c3 100644
>> --- a/drivers/lightnvm/pblk-write.c
>> +++ b/drivers/lightnvm/pblk-write.c
>> @@ -136,6 +136,7 @@ static void pblk_map_remaining(struct pblk *pblk, =
struct ppa_addr *ppa)
>> 		}
>> 	}
>>=20
>> +	line->w_err_gc->has_write_err =3D 1;
>> 	spin_unlock(&line->lock);
>> }
>>=20
>> @@ -279,6 +280,7 @@ static void pblk_end_io_write_meta(struct nvm_rq =
*rqd)
>> 	if (rqd->error) {
>> 		pblk_log_write_err(pblk, rqd);
>> 		pr_err("pblk: metadata I/O failed. Line %d\n", =
line->id);
>> +		line->w_err_gc->has_write_err =3D 1;
>> 	}
>>=20
>> 	sync =3D atomic_add_return(rqd->nr_ppas, &emeta->sync);
>> diff --git a/drivers/lightnvm/pblk.h b/drivers/lightnvm/pblk.h
>> index f8434a3..25ad026 100644
>> --- a/drivers/lightnvm/pblk.h
>> +++ b/drivers/lightnvm/pblk.h
>> @@ -89,12 +89,14 @@ struct pblk_sec_meta {
>> /* The number of GC lists and the rate-limiter states go together. =
This way the
>> * rate-limiter can dictate how much GC is needed based on resource =
utilization.
>> */
>> -#define PBLK_GC_NR_LISTS 3
>> +#define PBLK_GC_NR_LISTS 4
>>=20
>> enum {
>> -	PBLK_RL_HIGH =3D 1,
>> -	PBLK_RL_MID =3D 2,
>> -	PBLK_RL_LOW =3D 3,
>> +	PBLK_RL_OFF =3D 0,
>> +	PBLK_RL_WERR =3D 1,
>> +	PBLK_RL_HIGH =3D 2,
>> +	PBLK_RL_MID =3D 3,
>> +	PBLK_RL_LOW =3D 4
>> };
>>=20
>> #define pblk_dma_meta_size (sizeof(struct pblk_sec_meta) * =
PBLK_MAX_REQ_ADDRS)
>> @@ -278,6 +280,8 @@ struct pblk_rl {
>> 	int rb_user_active;
>> 	int rb_gc_active;
>>=20
>> +	atomic_t werr_lines;	/* Number of write error lines that =
needs gc */
>> +
>> 	struct timer_list u_timer;
>>=20
>> 	unsigned long long nr_secs;
>> @@ -311,6 +315,7 @@ enum {
>> 	PBLK_LINEGC_MID =3D 23,
>> 	PBLK_LINEGC_HIGH =3D 24,
>> 	PBLK_LINEGC_FULL =3D 25,
>> +	PBLK_LINEGC_WERR =3D 26
>> };
>>=20
>> #define PBLK_MAGIC 0x70626c6b /*pblk*/
>> @@ -412,6 +417,11 @@ struct pblk_smeta {
>> 	struct line_smeta *buf;		/* smeta buffer in persistent =
format */
>> };
>>=20
>> +struct pblk_w_err_gc {
>> +	int has_write_err;
>> +	__le64 *lba_list;
>> +};
>> +
>> struct pblk_line {
>> 	struct pblk *pblk;
>> 	unsigned int id;		/* Line number corresponds to =
the
>> @@ -457,6 +467,8 @@ struct pblk_line {
>>=20
>> 	struct kref ref;		/* Write buffer L2P references =
*/
>>=20
>> +	struct pblk_w_err_gc *w_err_gc;	/* Write error gc recovery =
metadata */
>> +
>> 	spinlock_t lock;		/* Necessary for invalid_bitmap =
only */
>> };
>>=20
>> @@ -488,6 +500,8 @@ struct pblk_line_mgmt {
>> 	struct list_head gc_mid_list;	/* Full lines ready to GC, mid =
isc */
>> 	struct list_head gc_low_list;	/* Full lines ready to GC, low =
isc */
>>=20
>> +	struct list_head gc_werr_list;  /* Write err recovery list */
>> +
>> 	struct list_head gc_full_list;	/* Full lines ready to GC, no =
valid */
>> 	struct list_head gc_empty_list;	/* Full lines close, all valid =
*/
>>=20
>> @@ -891,6 +905,9 @@ void pblk_rl_free_lines_dec(struct pblk_rl *rl, =
struct pblk_line *line,
>> 			    bool used);
>> int pblk_rl_is_limit(struct pblk_rl *rl);
>>=20
>> +void pblk_rl_werr_line_in(struct pblk_rl *rl);
>> +void pblk_rl_werr_line_out(struct pblk_rl *rl);
>> +
>> /*
>> * pblk sysfs
>> */
>> --
>> 2.7.4
>=20
> LGTM
>=20
> Reviewed-by: Javier Gonz=C3=A1lez <javier@cnexlabs.com>

--Apple-Mail=_033826BA-F62C-4A3B-8E14-C277C2A51A44
Content-Transfer-Encoding: 7bit
Content-Disposition: attachment; filename="signature.asc"
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Message signed with OpenPGP

-----BEGIN PGP SIGNATURE-----
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=OgYm
-----END PGP SIGNATURE-----

--Apple-Mail=_033826BA-F62C-4A3B-8E14-C277C2A51A44--