[PATCH v5 1/1] scsi: sas: skip opt_sectors when DMA reports no real optimization hint

[PATCH v5 1/1] scsi: sas: skip opt_sectors when DMA reports no real optimization hint

[Ionut Nechita (Wind River)]

From: Ionut Nechita <ionut.nechita@windriver.com>

sas_host_setup() unconditionally sets shost->opt_sectors from
dma_opt_mapping_size().  When the IOMMU is disabled or in passthrough
mode and no DMA ops provide an opt_mapping_size callback,
dma_opt_mapping_size() returns min(dma_max_mapping_size(), SIZE_MAX)
which equals dma_max_mapping_size() — a hard upper bound, not an
optimization hint.

On a Dell PowerEdge R750 with mpt3sas (Broadcom SAS3816, FW 33.15.00.00)
and intel_iommu=off the following values are observed:

  dma_opt_mapping_size()  = dma_max_mapping_size() (no real hint)
  shost->max_sectors      = 32767
  opt_sectors             = min(32767, huge >> 9) = 32767
  optimal_io_size         = 32767 << 9 = 16776704
                          → round_down(16776704, 4096) = 16773120

The SAS disk (SAMSUNG MZILT800HBHQ0D3) do not report an
Optimal Transfer Length in VPD page B0,so sdkp->opt_xfer_blocks remains 0.
sd_revalidate_disk() then uses min_not_zero(0, opt_sectors) = opt_sectors,
propagating the bogus value into the block device's optimal_io_size
(visible as OPT-IO = 16773120 in lsblk --topology).

mkfs.xfs picks up optimal_io_size and minimum_io_size and computes:

  swidth = 16773120 / 4096 = 4095
  sunit  = 8192 / 4096     = 2

Since 4095 % 2 != 0, XFS rejects the geometry:

  SB stripe unit sanity check failed

This makes it impossible to create XFS filesystems (e.g. for
/var/lib/docker) during system bootstrap.

Fix this by introducing a sas_dma_opt_sectors() helper that only returns
a non-zero opt_sectors when dma_opt_mapping_size() is strictly less than
dma_max_mapping_size(), indicating a genuine DMA optimization constraint
from an IOMMU or DMA ops backend.  The helper also rounds the value down
to a power of two so that filesystem geometry calculations always produce
clean results.  When the two DMA values are equal, no backend provided a
real hint, so opt_sectors stays at 0 ("no preference").

A WARN_ONCE guards against dma_opt_mapping_size() returning a value
larger than dma_max_mapping_size(), which would indicate a driver bug.
A zero check on opt guards against undefined behavior from
rounddown_pow_of_two(0).  The return value uses min_t(unsigned int, ...)
to avoid any potential overflow when shifting the size_t opt value down
to sectors.

Fixes: 4cbfca5f7750 ("scsi: scsi_transport_sas: cap shost opt_sectors according to DMA optimal limit")
Cc: stable@vger.kernel.org
Signed-off-by: Ionut Nechita <ionut.nechita@windriver.com>
---
 drivers/scsi/scsi_transport_sas.c | 52 ++++++++++++++++++++++++++++---
 1 file changed, 48 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c
index 12124f9d5ccd0..a5207caf8565e 100644
--- a/drivers/scsi/scsi_transport_sas.c
+++ b/drivers/scsi/scsi_transport_sas.c
@@ -27,6 +27,7 @@
 #include <linux/module.h>
 #include <linux/jiffies.h>
 #include <linux/err.h>
+#include <linux/log2.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/blkdev.h>
@@ -222,6 +223,50 @@ static int sas_bsg_initialize(struct Scsi_Host *shost, struct sas_rphy *rphy)
  * SAS host attributes
  */
 
+/**
+ * sas_dma_opt_sectors - derive opt_sectors from DMA optimal mapping size
+ * @dma_dev: device to query DMA parameters for
+ * @max_sectors: upper bound from the host adapter
+ *
+ * When the DMA layer reports a genuine optimization constraint (i.e.
+ * dma_opt_mapping_size() < dma_max_mapping_size()), convert it to a
+ * sector count, round it down to a power of two so that filesystem
+ * geometry calculations stay sane, and cap it at @max_sectors.
+ *
+ * When the two values are equal no backend provided a real hint and
+ * the function returns 0 ("no preference").  This happens when the
+ * IOMMU is disabled or in passthrough mode: dma_opt_mapping_size()
+ * falls back to min(SIZE_MAX, dma_max_mapping_size()) which equals
+ * dma_max_mapping_size().  Letting that value through would produce
+ * opt_sectors == max_sectors (e.g. 32767), leading to bogus
+ * optimal_io_size values that break mkfs.xfs stripe geometry.
+ *
+ * A WARN_ONCE guards against dma_opt_mapping_size() returning a value
+ * larger than dma_max_mapping_size(), which would indicate a driver bug.
+ */
+static unsigned int sas_dma_opt_sectors(struct device *dma_dev,
+					unsigned int max_sectors)
+{
+	size_t opt = dma_opt_mapping_size(dma_dev);
+	size_t max = dma_max_mapping_size(dma_dev);
+
+	if (WARN_ONCE(opt > max,
+		      "dma_opt_mapping_size (%zu) > dma_max_mapping_size (%zu)\n",
+		      opt, max))
+		return 0;
+
+	/* opt == max means no backend provided a real hint; see above. */
+	if (opt == max)
+		return 0;
+
+	if (!opt)
+		return 0;
+
+	opt = rounddown_pow_of_two(opt);
+
+	return min_t(unsigned int, opt >> SECTOR_SHIFT, max_sectors);
+}
+
 static int sas_host_setup(struct transport_container *tc, struct device *dev,
 			  struct device *cdev)
 {
@@ -239,10 +284,9 @@ static int sas_host_setup(struct transport_container *tc, struct device *dev,
 		dev_printk(KERN_ERR, dev, "fail to a bsg device %d\n",
 			   shost->host_no);
 
-	if (dma_dev->dma_mask) {
-		shost->opt_sectors = min_t(unsigned int, shost->max_sectors,
-				dma_opt_mapping_size(dma_dev) >> SECTOR_SHIFT);
-	}
+	if (dma_dev->dma_mask)
+		shost->opt_sectors =
+			sas_dma_opt_sectors(dma_dev, shost->max_sectors);
 
 	return 0;
 }
-- 
2.53.0

Re: [PATCH v5 1/1] scsi: sas: skip opt_sectors when DMA reports no real optimization hint

[John Garry]

On 20/03/2026 08:14, Ionut Nechita (Wind River) wrote:
> From: Ionut Nechita <ionut.nechita@windriver.com>
> 
> sas_host_setup() unconditionally sets shost->opt_sectors from
> dma_opt_mapping_size().  When the IOMMU is disabled or in passthrough
> mode and no DMA ops provide an opt_mapping_size callback,
> dma_opt_mapping_size() returns min(dma_max_mapping_size(), SIZE_MAX)
> which equals dma_max_mapping_size() — a hard upper bound, not an
> optimization hint.
> 
> On a Dell PowerEdge R750 with mpt3sas (Broadcom SAS3816, FW 33.15.00.00)
> and intel_iommu=off the following values are observed:
> 
>    dma_opt_mapping_size()  = dma_max_mapping_size() (no real hint)
>    shost->max_sectors      = 32767
>    opt_sectors             = min(32767, huge >> 9) = 32767
>    optimal_io_size         = 32767 << 9 = 16776704
>                            → round_down(16776704, 4096) = 16773120
> 
> The SAS disk (SAMSUNG MZILT800HBHQ0D3) do not report an
> Optimal Transfer Length in VPD page B0,so sdkp->opt_xfer_blocks remains 0.
> sd_revalidate_disk() then uses min_not_zero(0, opt_sectors) = opt_sectors,
> propagating the bogus value into the block device's optimal_io_size
> (visible as OPT-IO = 16773120 in lsblk --topology).
> 
> mkfs.xfs picks up optimal_io_size and minimum_io_size and computes:
> 
>    swidth = 16773120 / 4096 = 4095
>    sunit  = 8192 / 4096     = 2
> 
> Since 4095 % 2 != 0, XFS rejects the geometry:
> 
>    SB stripe unit sanity check failed
> 
> This makes it impossible to create XFS filesystems (e.g. for
> /var/lib/docker) during system bootstrap.
> 
> Fix this by introducing a sas_dma_opt_sectors() helper that only returns
> a non-zero opt_sectors when dma_opt_mapping_size() is strictly less than
> dma_max_mapping_size(), indicating a genuine DMA optimization constraint
> from an IOMMU or DMA ops backend.  The helper also rounds the value down
> to a power of two so that filesystem geometry calculations always produce
> clean results.  When the two DMA values are equal, no backend provided a
> real hint, so opt_sectors stays at 0 ("no preference").
> 
> A WARN_ONCE guards against dma_opt_mapping_size() returning a value
> larger than dma_max_mapping_size(), which would indicate a driver bug.
> A zero check on opt guards against undefined behavior from
> rounddown_pow_of_two(0).  The return value uses min_t(unsigned int, ...)
> to avoid any potential overflow when shifting the size_t opt value down
> to sectors.
> 
> Fixes: 4cbfca5f7750 ("scsi: scsi_transport_sas: cap shost opt_sectors according to DMA optimal limit")
> Cc: stable@vger.kernel.org
> Signed-off-by: Ionut Nechita <ionut.nechita@windriver.com>
> ---
>   drivers/scsi/scsi_transport_sas.c | 52 ++++++++++++++++++++++++++++---
>   1 file changed, 48 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c
> index 12124f9d5ccd0..a5207caf8565e 100644
> --- a/drivers/scsi/scsi_transport_sas.c
> +++ b/drivers/scsi/scsi_transport_sas.c
> @@ -27,6 +27,7 @@
>   #include <linux/module.h>
>   #include <linux/jiffies.h>
>   #include <linux/err.h>
> +#include <linux/log2.h>
>   #include <linux/slab.h>
>   #include <linux/string.h>
>   #include <linux/blkdev.h>
> @@ -222,6 +223,50 @@ static int sas_bsg_initialize(struct Scsi_Host *shost, struct sas_rphy *rphy)
>    * SAS host attributes
>    */
>   
> +/**

this is static, so really should not be a kerneldoc comment

> + * sas_dma_opt_sectors - derive opt_sectors from DMA optimal mapping size
> + * @dma_dev: device to query DMA parameters for
> + * @max_sectors: upper bound from the host adapter
> + *
> + * When the DMA layer reports a genuine optimization constraint (i.e.
> + * dma_opt_mapping_size() < dma_max_mapping_size()), convert it to a
> + * sector count, round it down to a power of two so that filesystem
> + * geometry calculations stay sane, and cap it at @max_sectors.
> + *
> + * When the two values are equal no backend provided a real hint and
> + * the function returns 0 ("no preference").  This happens when the
> + * IOMMU is disabled or in passthrough mode: dma_opt_mapping_size()
> + * falls back to min(SIZE_MAX, dma_max_mapping_size()) which equals
> + * dma_max_mapping_size().  Letting that value through would produce
> + * opt_sectors == max_sectors (e.g. 32767), leading to bogus
> + * optimal_io_size values that break mkfs.xfs stripe geometry.
> + *
> + * A WARN_ONCE guards against dma_opt_mapping_size() returning a value
> + * larger than dma_max_mapping_size(), which would indicate a driver bug.
> + */

too much is written here, it can be one paragraph

> +static unsigned int sas_dma_opt_sectors(struct device *dma_dev,
> +					unsigned int max_sectors)
> +{
> +	size_t opt = dma_opt_mapping_size(dma_dev);
> +	size_t max = dma_max_mapping_size(dma_dev);
> +
> +	if (WARN_ONCE(opt > max,
> +		      "dma_opt_mapping_size (%zu) > dma_max_mapping_size (%zu)\n",
> +		      opt, max))
> +		return 0;
> +

this can or should not happen, so it is not the drivers job to check it

> +	/* opt == max means no backend provided a real hint; see above. */
> +	if (opt == max)
> +		return 0;
> +
> +	if (!opt)
> +		return 0;

this can be combined with the previous check:

	if (!opt || opt == max)
		return;

> +
> +	opt = rounddown_pow_of_two(opt);
> +
> +	return min_t(unsigned int, opt >> SECTOR_SHIFT, max_sectors);

max_sectors can really be any value.

I would think that is is better to rounddown to a power-of-2 the min of 
opt sectors and max_sectors

> +}
> +
>   static int sas_host_setup(struct transport_container *tc, struct device *dev,
>   			  struct device *cdev)
>   {
> @@ -239,10 +284,9 @@ static int sas_host_setup(struct transport_container *tc, struct device *dev,
>   		dev_printk(KERN_ERR, dev, "fail to a bsg device %d\n",
>   			   shost->host_no);
>   
> -	if (dma_dev->dma_mask) {
> -		shost->opt_sectors = min_t(unsigned int, shost->max_sectors,
> -				dma_opt_mapping_size(dma_dev) >> SECTOR_SHIFT);
> -	}
> +	if (dma_dev->dma_mask)

I think that we don't need to declare dma_dev here, so can have:

static void sas_dma_setup_opt_sectors(struct Scsi_Host *shost)
{
	struct device *dma_dev = shost->dma_dev;

	if (!dma_dev->dma_mask)
		return;

	/* continue to evaluate opt sectors */
	...
}

> +		shost->opt_sectors =
> +			sas_dma_opt_sectors(dma_dev, shost->max_sectors);
>   
>   	return 0;
>   }