IIO: Alignment fixes part 4 - bounce buffers for the hard cases.
Jonathan Cameron
maintainercover IIO: Alignment fixes part 4 - bounce buffers for the hard cases.
From: Jonathan Cameron <[email protected]>
I finally got around to do a manual audit of all the calls to iio_push_to_buffers_with_timestamp() which has the somewhat odd requirements of:
1. 8 byte alignment of the provided buffer.
2. space for an 8 byte naturally aligned timestamp to be inserted at the end.
Unfortunately there were rather a lot of these left, but time to bite the bullet and clean them up.
As discussed previous in
https://lore.kernel.org/linux-iio/[email protected]/
it is not easy to fix the alignment issue without requiring a bounce buffer. This final part of the 4 sets of fixes is concerned with the cases where bounce buffers are the proposed solutions.
In these cases we have hardware that reads a prefix that we wish to drop. That makes it hard to directly read the data into the correct location.
Rather than implementing bouce buffers in each case, this set provides some magic in the core to handle them via a new function. iio_push_to_buffers_with_ts_na() - non aligned
Note this is totally untested as I don't have suitable hardware or emulation. I can fake something up in the dummy driver or via QEMU but I definitely want both eyes and testing on this series!
Jonathan Cameron (4):
iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned
case.
iio: adc: ti-adc108s102: Fix alignment of buffer pushed to iio
buffers.
iio: gyro: mpu3050: Fix alignment and size issues with buffers.
iio: imu: adis16400: Fix buffer alignment requirements.
drivers/iio/adc/ti-adc108s102.c | 11 +++++----- drivers/iio/gyro/mpu3050-core.c | 24 ++++++++++----------- drivers/iio/imu/adis16400.c | 20 ++++++++++++----- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 6 files changed, 76 insertions(+), 23 deletions(-)
Signature · 1 line
2.31.1
Jonathan Cameron
maintainerRFC PATCH 1/4 iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
From: Jonathan Cameron <[email protected]>
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to iio_push_to_buffers_with_timestamp(), there are a few places where leading data needs to be skipped over.
For these cases introduce a function that will allocate an appropriate sized and aligned bounce buffer (if not already allocated) and copy the unaligned data into that before calling iio_push_to_buffers_with_timestamp() on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.
--- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 3 files changed, 44 insertions(+)
drivers/iio/industrialio-buffer.c +36 -0
@@ -1727,6 +1727,42 @@ int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) }EXPORT_SYMBOL_GPL (iio_push_to_buffers); +/** +* iio_push_to_buffers_with_ts_na() - push to registered buffer, +* no alignment or space requirements. +* @indio_dev: iio_dev structure for device. +* @data: channel data excluding the timestamp. +* @data_sz: size of data. +* @timestamp: timestamp for the sample data. +* +* This special variant of iio_push_to_buffers_with_timetamp() does +* not require space for the timestamp, or 8 byte alignment of data. +* It does however require an allocation on first call and additional +* coppies on all calls, so should be avoided if possible +*/ +int iio_push_to_buffers_with_ts_na (struct iio_dev *indio_dev, +const void *data, +size_t data_sz, +int64_t timestamp) +{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz); +if (iio_dev_opaque->bounce_buffer_size != indio_dev->scan_bytes) { + iio_dev_opaque->bounce_buffer = +devm_krealloc (&indio_dev->dev, + iio_dev_opaque->bounce_buffer, + indio_dev->scan_bytes,GFP_KERNEL ); +if (!iio_dev_opaque) +return -ENOMEM ; + } +memcpy (iio_dev_opaque->bounce_buffer, data, data_sz); +return iio_push_to_buffers_with_timestamp (indio_dev, + iio_dev_opaque->bounce_buffer, + timestamp); +} +EXPORT_SYMBOL_GPL (iio_push_to_buffers_with_ts_na); +/** * iio_buffer_release() - Free a buffer's resources * @ref: Pointer to the kref embedded in the iio_buffer struct
include/linux/iio/buffer.h +4 -0
@@ -38,6 +38,10 @@ static inline int iio_push_to_buffers_with_timestamp(struct iio_dev *indio_dev,return iio_push_to_buffers (indio_dev, data); } +int iio_push_to_buffers_with_ts_na (struct iio_dev *indio_dev, +const void *data,size_t data_sz, +int64_t timestamp); +bool iio_validate_scan_mask_onehot (struct iio_dev *indio_dev,const unsigned long *mask);
include/linux/iio/iio-opaque.h +4 -0
@@ -19,6 +19,8 @@* @groupcounter: index of next attribute group * @legacy_scan_el_group: attribute group for legacy scan elements attribute group * @legacy_buffer_group: attribute group for legacy buffer attributes group +* @bounce_buffer: for devices that call iio_push_to_buffers_with_timestamp_na +* @bounce_buffer_size: size of currently allocate bounce buffer * @debugfs_dentry: device specific debugfs dentry * @cached_reg_addr: cached register address for debugfs reads * @read_buf: read buffer to be used for the initial reg read
@@ -38,6 +40,8 @@ struct iio_dev_opaque {int groupcounter;struct attribute_group legacy_scan_el_group;struct attribute_group legacy_buffer_group; +void *bounce_buffer; +size_t bounce_buffer_size;#if defined(CONFIG_DEBUG_FS )struct dentry *debugfs_dentry;unsigned cached_reg_addr;
Signature · 1 line
2.31.1
Jonathan Cameron
maintainerRFC PATCH 2/4 iio: adc: ti-adc108s102: Fix alignment of buffer pushed to iio buffers.
In reply to Jonathan CameronFrom: Jonathan Cameron <[email protected]>
Use the newly introduce iio_push_to_buffers_with_ts_na() function to ensure a bounce buffer is used to provide the required alignment and space padding needed by the IIO core which requires the timestamp is naturally aligned. There will be a performance cost to this change but it will ensure the driver works on platforms that do not support unaligned 8 byte assignements, and with consumer drivers that may assume natural alignment of the timestamp.
Issue found as part of an audit of all calls toiio_push_to_buffers_with_timestamp()
--- drivers/iio/adc/ti-adc108s102.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/iio/adc/ti-adc108s102.c +6 -5
@@ -75,9 +75,9 @@ struct adc108s102_state {* rx_buf: |XX|R0|R1|R2|R3|R4|R5|R6|R7|tt|tt|tt|tt| * * tx_buf: 8 channel read commands, plus 1 dummy command -* rx_buf: 1 dummy response, 8 channel responses, plus 64-bit timestamp +* rx_buf: 1 dummy response, 8 channel responses */ -__be16 rx_buf[13 ]____cacheline_aligned ; +__be16 rx_buf[9 ]____cacheline_aligned ;__be16 tx_buf[9 ]____cacheline_aligned ; };
@@ -149,9 +149,10 @@ static irqreturn_t adc108s102_trigger_handler(int irq, void *p) goto out_notify;/* Skip the dummy response in the first slot */ -iio_push_to_buffers_with_timestamp (indio_dev, - (u8 *)&st->rx_buf[1 ], -iio_get_time_ns (indio_dev)); +iio_push_to_buffers_with_ts_na (indio_dev, + &st->rx_buf[1 ], + st->ring_xfer.len -sizeof (st->rx_buf[1 ]), +iio_get_time_ns (indio_dev)); out_notify:iio_trigger_notify_done (indio_dev->trig);
Signature · 1 line
2.31.1
Jonathan Cameron
maintainerRFC PATCH 3/4 iio: gyro: mpu3050: Fix alignment and size issues with buffers.
In reply to Jonathan CameronFrom: Jonathan Cameron <[email protected]>
Fix a set of closely related issues.
1. When using fifo_values() there was not enough space for the timestamp to be inserted by iio_push_to_buffers_with_timestamp()
2. fifo_values() did not meet the alignment requirement of iio_push_to_buffers_with_timestamp()
3. hw_values did not meet the alignment requirement either.
1 and 2 fixed by using new iio_push_to_buffers_with_ts_na() which has no alignment or space padding requirements.
3 fixed by introducing a structure that makes the space and alignment requirements explicit.
--- drivers/iio/gyro/mpu3050-core.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-)
drivers/iio/gyro/mpu3050-core.c +11 -13
@@ -462,13 +462,10 @@ static irqreturn_t mpu3050_trigger_handler(int irq, void *p)struct iio_dev *indio_dev = pf->indio_dev;struct mpu3050 *mpu3050 =iio_priv (indio_dev);int ret; -/* -* Temperature 1*16 bits -* Three axes 3*16 bits -* Timestamp 64 bits (4*16 bits) -* Sum total 8*16 bits -*/ -__be16 hw_values[8 ]; +struct { +__be16 chans[4 ]; +s64 timestamp__aligned(8) ; + } scan;s64 timestamp;unsigned int datums_from_fifo =0 ;
@@ -563,9 +560,10 @@ static irqreturn_t mpu3050_trigger_handler(int irq, void *p) fifo_values[4]);/* Index past the footer (fifo_values[0]) and push */ -iio_push_to_buffers_with_timestamp (indio_dev, - &fifo_values[1 ], - timestamp); +iio_push_to_buffers_with_ts_na (indio_dev, + &fifo_values[1 ], +sizeof (__be16) *4 , + timestamp); fifocnt -= toread; datums_from_fifo++;
@@ -623,15 +621,15 @@ static irqreturn_t mpu3050_trigger_handler(int irq, void *p) goto out_trigger_unlock; } - ret =regmap_bulk_read (mpu3050->map,MPU3050_TEMP_H , &hw_values, -sizeof (hw_values)); + ret =regmap_bulk_read (mpu3050->map,MPU3050_TEMP_H , scan.chans, +sizeof (scan.chans));if (ret) {dev_err (mpu3050->dev,"error reading axis data\n" ); goto out_trigger_unlock; } -iio_push_to_buffers_with_timestamp (indio_dev, hw_values, timestamp); +iio_push_to_buffers_with_timestamp (indio_dev, &scan, timestamp); out_trigger_unlock:mutex_unlock (&mpu3050->lock);
Signature · 1 line
2.31.1
Jonathan Cameron
maintainerRFC PATCH 4/4 iio: imu: adis16400: Fix buffer alignment requirements.
In reply to Jonathan CameronFrom: Jonathan Cameron <[email protected]>
iio_push_to_buffers_with_timestamp() requires that the buffer is 8 byte alignment to ensure an inserted timestamp is naturally aligned.
This requirement was not met here when burst mode is in use beause of a leading u16. Use the new iio_push_to_buffers_with_ts_na() function that has more relaxed requirements.
It is somewhat complex to access that actual data length, but a safe bound can be found by using scan_bytes - sizeof(timestamp) so that is used in this path.
More efficient approaches exist, but this ensure correctness at the cost of using a bounce buffer.
--- drivers/iio/imu/adis16400.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-)
drivers/iio/imu/adis16400.c +15 -5
@@ -663,13 +663,23 @@ static irqreturn_t adis16400_trigger_handler(int irq, void *p)spi_setup (st->adis.spi); } -if (st->variant->flags &ADIS16400_BURST_DIAG_STAT ) +if (st->variant->flags &ADIS16400_BURST_DIAG_STAT ) { buffer = adis->buffer +sizeof (u16); -else - buffer = adis->buffer; +/* +* The size here is always larger than, or equal to the true +* size of the channel data. This may result in a larger copy +* than necessary, but as the target buffer will be +* buffer->scan_bytes this will be safe. +*/ +iio_push_to_buffers_with_ts_na (indio_dev, buffer, + indio_dev->scan_bytes -sizeof (pf->timestamp), + pf->timestamp); + }else { +iio_push_to_buffers_with_timestamp (indio_dev, + adis->buffer, + pf->timestamp); + } -iio_push_to_buffers_with_timestamp (indio_dev, buffer, - pf->timestamp);iio_trigger_notify_done (indio_dev->trig);
Signature · 1 line
2.31.1
Re: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
In reply to Jonathan CameronQuoted 14 lines · Jonathan Cameron
From: Jonathan Cameron <[email protected]>
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a few places where leading data needs to be skipped over.For these cases introduce a function that will allocate an appropriate sized and aligned bounce buffer (if not already allocated) and copy the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.
...
Quoted 7 lines
+/**
+ *iio_push_to_buffers_with_ts_na()- push to registered buffer,
+ * no alignment or space requirements.+ * @indio_dev: iio_dev structure for device. + * @data: channel data excluding the timestamp. + * @data_sz: size of data. + * @timestamp: timestamp for the sample data.
Quoted 5 lines
+ *
+ * This special variant ofiio_push_to_buffers_with_timetamp()does
+ * not require space for the timestamp, or 8 byte alignment of data.
+ * It does however require an allocation on first call and additional
+ * coppies on all calls, so should be avoided if possible
copies
Quoted 1 line
+ */
I do not like the _na part in the name (My first impression was with a Timestamp that was not available, what?!). Can we spell it better?
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1754
Quoted 13 lines
drivers/iio/industrialio-buffer.c +12
-0+int iio_push_to_buffers_with_ts_na (struct iio_dev *indio_dev, +const void *data, +size_t data_sz, +int64_t timestamp) +{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz); +if (iio_dev_opaque->bounce_buffer_size != indio_dev->scan_bytes) { + iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer,
Oh la la, foo = realloc(foo, ...) is 101 type of mistakes. Please, don't do this.
Quoted 9 lines
drivers/iio/industrialio-buffer.c +9
-0+ indio_dev->scan_bytes, GFP_KERNEL); +if (!iio_dev_opaque) +return -ENOMEM ; + } +memcpy (iio_dev_opaque->bounce_buffer, data, data_sz); +return iio_push_to_buffers_with_timestamp (indio_dev, + iio_dev_opaque->bounce_buffer, + timestamp); +}
Signature · 2 lines
With Best Regards, Andy Shevchenko
Re: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
Quoted 1 line · Andy Shevchenko
On Sat, May 1, 2021 at 8:28 PM Jonathan Cameron <[email protected]> wrote:
...
Quoted 9 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1754
Quoted 3 lines
drivers/iio/industrialio-buffer.c +3
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer,Oh la la, foo =
realloc(foo, ...)is 101 type of mistakes. Please, don't do this.Quoted 2 lines
drivers/iio/industrialio-buffer.c +2
-0+ indio_dev->scan_bytes, GFP_KERNEL); +if (!iio_dev_opaque)
And this check...
Quoted 2 lines
Quoted 2 lines
(quoted diff) +2
-0+ return -ENOMEM; + }
Signature · 2 lines
With Best Regards, Andy Shevchenko
RE: [RFC PATCH 4/4] iio: imu: adis16400: Fix buffer alignment requirements.
In reply to Jonathan CameronQuoted 61 lines
From: Jonathan Cameron <[email protected]>
Sent: Saturday, May 1, 2021 7:25 PM
To: [email protected]
Cc: Jonathan Cameron <[email protected]>; Sa, Nuno <[email protected]>
Subject: [RFC PATCH 4/4] iio: imu: adis16400: Fix buffer alignment requirements.From: Jonathan Cameron <[email protected]>
iio_push_to_buffers_with_timestamp()requires that the buffer is 8 byte alignment to ensure an inserted timestamp is naturally aligned.This requirement was not met here when burst mode is in use beause of a leading u16. Use the new
iio_push_to_buffers_with_ts_na()function
that has more relaxed requirements.It is somewhat complex to access that actual data length, but a safe bound can be found by using
scan_bytes- sizeof(timestamp) so that
is used in this path.More efficient approaches exist, but this ensure correctness at the cost of using a bounce buffer.
Fixes 5075e0720d93 ("iio: imu: adis: generalize burst mode support")Signed-off-by Jonathan Cameron <[email protected]>Cc Nuno Sa <[email protected]>--- drivers/iio/imu/adis16400.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-)drivers/iio/imu/adis16400.c +3
-3@@ -663,13 +663,23 @@ static irqreturn_t adis16400_trigger_handler(int irq, void *p) spi_setup(st->adis.spi); } -if (st->variant->flags &ADIS16400_BURST_DIAG_STAT ) +if (st->variant->flags &ADIS16400_BURST_DIAG_STAT ) { buffer = adis->buffer +sizeof (u16); -else - buffer = adis->buffer; +/* +* The size here is always larger than, or equal to the true
drivers/iio/imu/adis16400.c +1
-0+* size of the channel data. This may result in a larger copy
drivers/iio/imu/adis16400.c +5
-0+* than necessary, but as the target buffer will be +* buffer->scan_bytes this will be safe. +*/ + iio_push_to_buffers_with_ts_na(indio_dev, buffer, + indio_dev->scan_bytes -
sizeof(pf->timestamp),
drivers/iio/imu/adis16400.c static irqreturn_t adis16400_trigger_handler(int irq, void *p) · line 681
Quoted 6 lines
drivers/iio/imu/adis16400.c +6
-0+ pf->timestamp); + }else { +iio_push_to_buffers_with_timestamp (indio_dev, + adis->buffer, + pf->timestamp); + }
Hi Jonathan,
This looks good to me although I think there's some stuff to care in ' iio_push_to_buffers_with_ts_na()'. However, for this patch alone:
I will also check if I find any HW to test this...
- Nuno Sá
Sa, Nuno
maintainerRE: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
In reply to Jonathan CameronQuoted 51 lines
From: Jonathan Cameron <[email protected]>
Sent: Saturday, May 1, 2021 7:25 PM
To: [email protected]
Cc: Jonathan Cameron <[email protected]>
Subject: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.From: Jonathan Cameron <[email protected]>
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a few places where leading data needs to be skipped over.For these cases introduce a function that will allocate an appropriate sized and aligned bounce buffer (if not already allocated) and copy the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.Signed-off-by Jonathan Cameron <[email protected]>--- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 3 files changed, 44 insertions(+)drivers/iio/industrialio- renamed from drivers/iio/industrialio-buffer.c +0
-0
buffer.c
index9a8e16c7e9af..818dfaa73665100644drivers/iio/industrialio-buffer.c +12
-0@@ -1727,6 +1727,42 @@ int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) }EXPORT_SYMBOL_GPL (iio_push_to_buffers); +/** +* iio_push_to_buffers_with_ts_na() - push to registered buffer, +* no alignment or space requirements. +* @indio_dev: iio_dev structure for device. +* @data: channel data excluding the timestamp. +* @data_sz: size of data. +* @timestamp: timestamp for the sample data. +* +* This special variant of iio_push_to_buffers_with_timetamp() does +* not require space for the timestamp, or 8 byte alignment of data. +* It does however require an allocation on first call and additional +* coppies on all calls, so should be avoided if possible
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1752
Quoted 10 lines
drivers/iio/industrialio-buffer.c +9
-0+*/ +int iio_push_to_buffers_with_ts_na (struct iio_dev *indio_dev, +const void *data, +size_t data_sz, +int64_t timestamp) +{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz);
I'm not really sure about this. Is it really a good idea to silently truncate the data if some erroneous driver calls this with data_sz > scan_bytes? (I don't think it's ever valid for data_sz to be bigger than scan_bytes) We might be discarding data that userland actually requested. So, I'm not sure if it would not be better to be strict here and return error, so that driver developers could catch this early on...
data_sz < scan_bytes is also probably not valid as that might happen if some drivers call this without the scan elements properly aligned which will definitely be problematic... Naturally in this case we would have to take into account timestamp possible alignment fill bytes as it could be possible (and likely) that data_sz won't include those bits...
Alternatively, we could be consistent with 'iio_push_to_buffers_with_timestamp()' and assume drivers play along in terms of alignment and size
(so, removing the data_sz parameter)...
My point is, if we are taking the data_sz parameter, maybe we should be pedantic about it...
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1754
Quoted 2 lines
drivers/iio/industrialio-buffer.c +1
-0+if (iio_dev_opaque->bounce_buffer_size != indio_dev- >scan_bytes) {
Where do we set ' bounce_buffer_size '? I guess we do not want to get here all the time...
Quoted 5 lines
drivers/iio/industrialio-buffer.c +4
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer, + indio_dev->scan_bytes,
GFP_KERNEL);
I'm also not a big fan of realloc so I prefer to avoid it unless really needed... Here, maybe we could just allocate this once during device/buffer initialization and allocate it for the maximum possible size, i.e all scan elements enabled... That could also make this API simpler to the point of actually be inlined...
- Nuno Sá
Jonathan Cameron
maintainerRe: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
In reply to Andy ShevchenkoQuoted 10 lines · On Sat, 1 May 2021 22:27:22 +0300 Andy Shevchenko <[email protected]>…
Quoted 1 line · Andy Shevchenko
On Sat, May 1, 2021 at 8:28 PM Jonathan Cameron <[email protected]> wrote:
...
Quoted 4 lines
Quoted 3 lines
drivers/iio/industrialio-buffer.c +3
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer,
Quoted 7 lines
Quoted 5 lines
Oh la la, foo =
realloc(foo, ...)is 101 type of mistakes. Please, don't do this.Quoted 2 lines
drivers/iio/industrialio-buffer.c +2
-0+ indio_dev->scan_bytes, GFP_KERNEL); +if (!iio_dev_opaque)And this check...
Yup. Thoroughly broken. Hohum, but basic idea conveyed however bad the rfc was :(
Quoted 4 lines
Quoted 3 lines
Quoted 3 lines
(quoted diff) +2
-0+ return -ENOMEM; + }
Jonathan Cameron
maintainerRe: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
In reply to Andy ShevchenkoQuoted 31 lines · On Sat, 1 May 2021 22:25:55 +0300 Andy Shevchenko <[email protected]>…
Quoted 14 lines · Jonathan Cameron
From: Jonathan Cameron <[email protected]>
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a few places where leading data needs to be skipped over.For these cases introduce a function that will allocate an appropriate sized and aligned bounce buffer (if not already allocated) and copy the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur....
Quoted 7 lines
+/**
+ *iio_push_to_buffers_with_ts_na()- push to registered buffer,
+ * no alignment or space requirements.+ * @indio_dev: iio_dev structure for device. + * @data: channel data excluding the timestamp. + * @data_sz: size of data. + * @timestamp: timestamp for the sample data.Quoted 5 lines
+ *
+ * This special variant ofiio_push_to_buffers_with_timetamp()does
+ * not require space for the timestamp, or 8 byte alignment of data.
+ * It does however require an allocation on first call and additional
+ * coppies on all calls, so should be avoided if possible
Quoted 1 line
copies
One day I'll remember to actually spell check *sigh*
Quoted 5 lines
Quoted 2 lines
+ */
I do not like the
_napart in the name (My first impression was with a Timestamp that was not available, what?!). Can we spell it better?
I struggled with the naming. Ideally we'd have started with this as the iio_push_to_buffers_with_timestamp() and had
an _aligned version for the existing case.
Perhaps spend the characters and just make it_with_ts_unaligned()
Quoted 11 lines
Quoted 10 lines
drivers/iio/industrialio-buffer.c +9
-0+int iio_push_to_buffers_with_ts_na (struct iio_dev *indio_dev, +const void *data, +size_t data_sz, +int64_t timestamp) +{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz); +if (iio_dev_opaque->bounce_buffer_size != indio_dev->scan_bytes) {
Quoted 6 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1754
Quoted 3 lines
drivers/iio/industrialio-buffer.c +3
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer,Oh la la, foo =
realloc(foo, ...)is 101 type of mistakes. Please, don't do this.
For realloc I'd agree because if the new allocation fails we'd just have lost the pointer, but with a managed case, I think we'll leave the original pointer alone from the point of view of the devm_ cleanup.
The only exit paths of interest in devm_krealloc() are the ones where we are trying to allocate a new larger object (otherwise it either does nothing or it is just a call to devm_kmalloc().
The one on failure to find the original managed resource, so can't loose it because it wasn't there.
Am I missing something?
Quoted 4 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1757
Quoted 4 lines
drivers/iio/industrialio-buffer.c +3
-0+ indio_dev->scan_bytes, GFP_KERNEL); +if (!iio_dev_opaque) +return -ENOMEM ;
As you observed this is clearly garbage. I should have sat on this patch for a day and at least reread it or ideally done some testing.
Failing to set the bounce_buffer_size() definitely doesn't help either..
Quoted 7 lines
Quoted 6 lines
drivers/iio/industrialio-buffer.c +6
-0+ } +memcpy (iio_dev_opaque->bounce_buffer, data, data_sz); +return iio_push_to_buffers_with_timestamp (indio_dev, + iio_dev_opaque->bounce_buffer, + timestamp); +}
Jonathan Cameron
maintainerRe: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
In reply to Sa, NunoQuoted 56 lines · Nuno
Quoted 56 lines
From: Jonathan Cameron <[email protected]>
Sent: Saturday, May 1, 2021 7:25 PM
To: [email protected]
Cc: Jonathan Cameron <[email protected]>
Subject: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.From: Jonathan Cameron <[email protected]>
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a few places where leading data needs to be skipped over.For these cases introduce a function that will allocate an appropriate sized and aligned bounce buffer (if not already allocated) and copy the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.Signed-off-by Jonathan Cameron <[email protected]>--- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 3 files changed, 44 insertions(+)drivers/iio/industrialio- renamed from drivers/iio/industrialio-buffer.c +0
-0
buffer.c
index9a8e16c7e9af..818dfaa73665100644drivers/iio/industrialio-buffer.c +17
-0@@ -1727,6 +1727,42 @@ int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) }EXPORT_SYMBOL_GPL (iio_push_to_buffers); +/** +* iio_push_to_buffers_with_ts_na() - push to registered buffer, +* no alignment or space requirements. +* @indio_dev: iio_dev structure for device. +* @data: channel data excluding the timestamp. +* @data_sz: size of data. +* @timestamp: timestamp for the sample data. +* +* This special variant of iio_push_to_buffers_with_timetamp() does +* not require space for the timestamp, or 8 byte alignment of data. +* It does however require an allocation on first call and additional +* coppies on all calls, so should be avoided if possible +*/ +int iio_push_to_buffers_with_ts_na (struct iio_dev *indio_dev, +const void *data, +size_t data_sz, +int64_t timestamp)
Quoted 8 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1750
Quoted 5 lines
drivers/iio/industrialio-buffer.c +4
-0+{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz);I'm not really sure about this. Is it really a good idea to silently truncate the data if some erroneous driver calls this with
data_sz>scan_bytes? (I don't think it's ever valid fordata_szto be bigger thanscan_bytes)
It's not probably not. This is just a fairly nasty way of papering over anyone doing that.
Quoted 3 lines
We might be discarding data that userland actually requested. So, I'm not sure if it would not be better to be strict here and return error, so that driver developers could catch this early on...
I'm fairly sure we'll never be discarding data userspace requested because scan_bytes will be the maximum possible and is the size of the kfifo element. Any more data that comes in is going to be dropped anyway assuming it ends up in a kfifo_put()
I wondered if we wanted to enforce this? It would not seem totally silly for a driver that wasn't bothering to track the size of it's enabled channels to provide data_sz == maximum size it ever needs.
Docs should definitely say if that's valid however, and it's definitely less than optimal because we may copy garbage that will then be overwritten by the timestamp.
Quoted 6 lines
data_sz<scan_bytesis also probably not valid as that might happen if some drivers call this without the scan elements properly aligned which will definitely be problematic... Naturally in this case we would have to take into account timestamp possible alignment fill bytes as it could be possible (and likely) thatdata_szwon't include those bits...
data_sz < scan_bytes is intentionally valid (and indeed I made one of the drivers do this). Given we are bouncing anyway, why insist the driver provides space for padding + timestamp? That can be done for the bounce buffer.
Quoted 5 lines
Alternatively, we could be consistent with '
iio_push_to_buffers_with_timestamp()' and assume drivers play along in terms of alignment and size
(so, removing thedata_szparameter)...
Quoted 7 lines
My point is, if we are taking the
data_szparameter, maybe we should be pedantic about it...drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1754
Quoted 2 lines
drivers/iio/industrialio-buffer.c +1
-0+if (iio_dev_opaque->bounce_buffer_size != indio_dev- >scan_bytes) {Where do we set '
bounce_buffer_size'? I guess we do not want to get here all the time...
:( oops
Quoted 9 lines
Quoted 5 lines
drivers/iio/industrialio-buffer.c +4
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer, + indio_dev->scan_bytes,
GFP_KERNEL);I'm also not a big fan of realloc so I prefer to avoid it unless really needed... Here, maybe we could just allocate this once during device/buffer initialization and allocate it for the maximum possible size, i.e all scan elements enabled...
We can but to do that requires more invasive changes. Perhaps not horrible ones, but we'd need to add stuff to caller drivers to configure the bounce buffer, or some flag to see they needed to enable support for this. Only 3 drivers out of all those in IIO are currently using this code (might be a few others that could take advantage but not many - so need to keep it localized).
I suppose we could have aniio_unaligned_push_to_buffers_prepare() to be called from the update_scan_mode() callback (typed to be useable as that callback if nothing else to be done in the relevant drivers.
Quoted 1 line
That could also make this API simpler to the point of actually be inlined...
Can't make it inline without making the bounce buffer visible to the driver and I don't want the driver to be able to see it as it adds another place for things to be done wrongly in drivers.
Any driver doing this is having to deal with a bounce buffer anyway, so the indirection doesn't worry me too much.
Quoted 2 lines
- Nuno Sá
Thanks,
Jonathan
Sa, Nuno
maintainerRE: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
Quoted 94 lines
-----Original Message-----
From: Jonathan Cameron <[email protected]>
Sent: Sunday, May 2, 2021 6:09 PM
To: Sa, Nuno <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: Re: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 7 lines · Nuno
Quoted 7 lines
From: Jonathan Cameron <[email protected]>
Sent: Saturday, May 1, 2021 7:25 PM
To: [email protected]
Cc: Jonathan Cameron <[email protected]>
Subject: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 5 lines
Quoted 5 lines
From: Jonathan Cameron <[email protected]>
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a few placeswhere
Quoted 3 lines
Quoted 3 lines
leading data needs to be skipped over.
For these cases introduce a function that will allocate an
appropriate
Quoted 7 lines
Quoted 7 lines
sized and aligned bounce buffer (if not already allocated) and copy the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.Signed-off-by Jonathan CameronQuoted 8 lines
Review
Quoted 8 lines
--- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 3 files changed, 44 insertions(+)(quoted diff) +0
-0b/drivers/iio/industrialio-
Quoted 9 lines
Quoted 9 lines
buffer.c
index9a8e16c7e9af..818dfaa73665100644drivers/iio/industrialio-buffer.c +0
-0@@ -1727,6 +1727,42 @@ int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) }EXPORT_SYMBOL_GPL (iio_push_to_buffers);Quoted 9 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data)
Quoted 9 lines
drivers/iio/industrialio-buffer.c +9
-0+/** +* iio_push_to_buffers_with_ts_na() - push to registered buffer, +* no alignment or space requirements. +* @indio_dev: iio_dev structure for device. +* @data: channel data excluding the timestamp. +* @data_sz: size of data. +* @timestamp: timestamp for the sample data. +* +* This special variant of iio_push_to_buffers_with_timetamp() does
Quoted 1 line
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1740
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+* not require space for the timestamp, or 8 byte alignment of data.
Quoted 5 lines
Quoted 5 lines
drivers/iio/industrialio-buffer.c +5
-0+* It does however require an allocation on first call and additional +* coppies on all calls, so should be avoided if possible +*/ +int iio_push_to_buffers_with_ts_na(struct iio_dev *indio_dev, + const void *data,Quoted 8 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1750
Quoted 7 lines
drivers/iio/industrialio-buffer.c +6
-0+ size_t data_sz, + int64_t timestamp) +{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz);I'm not really sure about this. Is it really a good idea to silently
truncate
Quoted 1 line
the data if some erroneous driver calls this with
data_sz>
scan_bytes?Quoted 1 line
(I don't think it's ever valid for
data_szto be bigger thanscan_bytes)It's not probably not. This is just a fairly nasty way of papering over anyone doing that.
Quoted 2 lines
We might be discarding data that userland actually requested. So, I'm not sure if it would not be better to be strict here and return error,
so that
Quoted 8 lines
Quoted 1 line
driver developers could catch this early on...
I'm fairly sure we'll never be discarding data userspace requested because
scan_byteswill be the maximum possible and is the size of the
kfifo element. Any more data that comes in is going to be dropped anyway
assuming it ends up in akfifo_put()
I think it depends. Think on a totally erroneous driver that does not track enabled elements and just pushes all elements to IIO buffers. If userspace just requested, let's say, the last 2 elements, I believe those would be definitely discarded by this... More subtle things could also happen I guess... In this case, we can arguably say that the driver is completely broken and while I'm fairly sure such a scenario would never pass through revision, I think that if we make things more strict, we might minimize the chances of such patches to appear in the first place (unless someone is sending patches without testing them at all).
Quoted 15 lines
I wondered if we wanted to enforce this? It would not seem totally silly for a driver that wasn't bothering to track the size of it's enabled channels to provide
data_sz== maximum size it ever needs.
Docs should definitely say if that's valid however, and it's definitely less than optimal because we may copy garbage that will then be overwritten by
the timestamp.Quoted 6 lines
data_sz<scan_bytesis also probably not valid as that might happen if some drivers call this without the scan elements properly aligned which will definitely be problematic... Naturally in this case we would have to take into account timestamp possible alignment fill bytes as it could be possible (and likely) thatdata_szwon't include those bits...
Quoted 4 lines
data_sz<scan_bytesis intentionally valid (and indeed I made one of the drivers do this). Given we are bouncing anyway, why insist the driver provides space for padding + timestamp? That can be done for the bounce buffer.
What I meant was that if data_sz < scan_bytes - timestamp - padding, then the driver is most likely not guaranteeing properly aligned elements. I was not that clear :)
Anyways, I still feel that if we are taking the data_sz parameter on the API, we should handle it more strictly. But I do not feel too strong about this (as you said, It's not expected to exist too much users of this) and in the end of the day it still behaves the same way as 'iio_push_to_buffers_with_timestamp()'...
Quoted 30 lines
Quoted 2 lines
Alternatively, we could be consistent with
'
iio_push_to_buffers_with_timestamp()'Quoted 4 lines
and assume drivers play along in terms of alignment and size (so, removing the
data_szparameter)...My point is, if we are taking the
data_szparameter, maybe weshould be pedantic
Quoted 6 lines
about it...
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1754
Quoted 2 lines
drivers/iio/industrialio-buffer.c +1
-0+if (iio_dev_opaque->bounce_buffer_size != indio_dev- >scan_bytes) {Where do we set '
bounce_buffer_size'? I guess we do not want to get here all the time...:( oops
Quoted 7 lines
Quoted 5 lines
drivers/iio/industrialio-buffer.c +4
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer, + indio_dev->scan_bytes,
GFP_KERNEL);I'm also not a big fan of realloc so I prefer to avoid it unless really
needed...
Quoted 1 line
Here, maybe we could just allocate this once during device/buffer
initialization
Quoted 1 line
and allocate it for the maximum possible size, i.e all scan elements
enabled...
Quoted 9 lines
We can but to do that requires more invasive changes. Perhaps not horrible ones,
but we'd need to add stuff to caller drivers to configure the bounce buffer, or
some flag to see they needed to enable support for this. Only 3 drivers out of
all those in IIO are currently using this code (might be a few others that could
take advantage but not many - so need to keep it localized).
I was not thinking that way. I was just thinking in always allocating the bounce buffer. But I do agree that it is less than ideal to allocate something that will almost never be used...
Quoted 6 lines
I suppose we could have an
iio_unaligned_push_to_buffers_prepare()to be called from theupdate_scan_mode()
callback (typed to be useable as that callback if nothing else to be done in
the relevant drivers.
It could also be an option... If we want to not use realloc, we would have to manually control the lifespan of the buffer and free it in the disable path. But I guess that might feel like too much handling...
Anyways, maybe sticking with realloc, we could just use a more normal pattern like:
tmp = devm_realloc(bounce_fuffer, ...);
if (!tmp)
error;
bounce_buffer = tmp;
- Nuno Sá
Sa, Nuno
maintainerRE: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
Quoted 240 lines
-----Original Message-----
From: Sa, Nuno <[email protected]>
Sent: Monday, May 3, 2021 9:15 AM
To: Jonathan Cameron <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: RE: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 22 lines
-----Original Message-----
From: Jonathan Cameron <[email protected]>
Sent: Sunday, May 2, 2021 6:09 PM
To: Sa, Nuno <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: Re: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 9 lines · Nuno
Quoted 9 lines
From: Jonathan Cameron <[email protected]>
Sent: Saturday, May 1, 2021 7:25 PM
To: [email protected]
Cc: Jonathan Cameron <[email protected]>
Subject: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.From: Jonathan Cameron <[email protected]>
Quoted 9 lines
Quoted 3 lines
Quoted 3 lines
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a few placeswhere
Quoted 3 lines
Quoted 3 lines
leading data needs to be skipped over.
For these cases introduce a function that will allocate an
appropriate
Quoted 1 line
Quoted 1 line
sized and aligned bounce buffer (if not already allocated) and
copy
Quoted 25 lines
Quoted 6 lines
Quoted 6 lines
the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.Signed-off-by Jonathan CameronQuoted 8 lines
Review
Quoted 8 lines
--- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 3 files changed, 44 insertions(+)(quoted diff) +0
-0b/drivers/iio/industrialio-
Quoted 9 lines
Quoted 9 lines
buffer.c
index9a8e16c7e9af..818dfaa73665100644drivers/iio/industrialio-buffer.c +0
-0@@ -1727,6 +1727,42 @@ int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) }EXPORT_SYMBOL_GPL (iio_push_to_buffers);Quoted 2 lines
Quoted 2 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1732
Quoted 2 lines
drivers/iio/industrialio-buffer.c +2
-0+/** +* iio_push_to_buffers_with_ts_na() - push to registered buffer,
Quoted 11 lines
Quoted 7 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data)
Quoted 7 lines
drivers/iio/industrialio-buffer.c +7
-0+* no alignment or space requirements. +* @indio_dev: iio_dev structure for device. +* @data: channel data excluding the timestamp. +* @data_sz: size of data. +* @timestamp: timestamp for the sample data. +* +* This special variant of iio_push_to_buffers_with_timetamp() does
Quoted 1 line
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1740
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+* not require space for the timestamp, or 8 byte alignment of data.
Quoted 1 line
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+* It does however require an allocation on first call and additional
Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
drivers/iio/industrialio-buffer.c +4
-0+* coppies on all calls, so should be avoided if possible +*/ +int iio_push_to_buffers_with_ts_na(struct iio_dev *indio_dev, + const void *data,Quoted 8 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1750
Quoted 7 lines
drivers/iio/industrialio-buffer.c +6
-0+ size_t data_sz, + int64_t timestamp) +{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz);I'm not really sure about this. Is it really a good idea to silently
Quoted 4 lines
truncate
Quoted 1 line
the data if some erroneous driver calls this with
data_sz>
scan_bytes?Quoted 1 line
(I don't think it's ever valid for
data_szto be bigger than
scan_bytes)Quoted 5 lines
It's not probably not. This is just a fairly nasty way of papering over anyone doing that.
Quoted 1 line
We might be discarding data that userland actually requested. So,
I'm
Quoted 10 lines
Quoted 1 line
not sure if it would not be better to be strict here and return error,
so that
Quoted 1 line
driver developers could catch this early on...
I'm fairly sure we'll never be discarding data userspace requested because
scan_byteswill be the maximum possible and is the size of the
kfifo element. Any more data that comes in is going to be dropped anyway
assuming it ends up in akfifo_put()I think it depends. Think on a totally erroneous driver that does not track
enabled elements and just pushes all elements to IIO buffers. If userspace
just requested, let's say, the last 2 elements, I believe those would be definitely discarded by this... More subtle things could also happen I guess...
In this case, we can arguably say that the driver is completely broken and while
I'm fairly sure such a scenario would never pass through revision, I think that
if we make things more strict, we might minimize the chances of such patches
to appear in the first place (unless someone is sending patches without testing them at all).Quoted 4 lines
I wondered if we wanted to enforce this? It would not seem totally silly for a driver that wasn't bothering to track the size of it's enabled channels to provide
data_sz== maximum size it ever needs. Docs should definitely say if that's valid however, and it's definitelyless
Quoted 6 lines
than optimal because we may copy garbage that will then be overwritten by
the timestamp.Quoted 2 lines
data_sz<scan_bytesis also probably not valid as that mighthappen
Quoted 2 lines
Quoted 2 lines
if some drivers call this without the scan elements properly aligned which will definitely be problematic... Naturally in this case we
would
Quoted 1 line
Quoted 1 line
have to take into account timestamp possible alignment fill bytes as
it
Quoted 1 line
Quoted 1 line
could be possible (and likely) that
data_szwon't include thosebits...
Quoted 3 lines
data_sz<scan_bytesis intentionally valid (and indeed I made one of the drivers do this). Given we are bouncing anyway, why insistthe
Quoted 1 line
driver provides space for padding + timestamp? That can be done
for
Quoted 1 line
the bounce buffer.
What I meant was that if
data_sz<scan_bytes- timestamp - padding, then
the driver is most likely not guaranteeing properly aligned elements. I was
not that clear :)Anyways, I still feel that if we are taking the
data_szparameter on the API,
we should handle it more strictly. But I do not feel too strong about this (as
you said, It's not expected to exist too much users of this) and in the end
of the day it still behaves the same way as 'iio_push_to_buffers_with_timestamp()'...Quoted 13 lines
Quoted 2 lines
Alternatively, we could be consistent with
'
iio_push_to_buffers_with_timestamp()'Quoted 4 lines
and assume drivers play along in terms of alignment and size (so, removing the
data_szparameter)...My point is, if we are taking the
data_szparameter, maybe weshould be pedantic
Quoted 5 lines
about it...
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1754
Quoted 2 lines
drivers/iio/industrialio-buffer.c +1
-0+if (iio_dev_opaque->bounce_buffer_size != indio_dev- >scan_bytes) {Where do we set '
bounce_buffer_size'? I guess we do not wantto
Quoted 17 lines
Quoted 1 line
get here all the time...
:( oops
Quoted 7 lines
Quoted 5 lines
drivers/iio/industrialio-buffer.c +4
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer, + indio_dev->scan_bytes,
GFP_KERNEL);I'm also not a big fan of realloc so I prefer to avoid it unless really
needed...
Quoted 1 line
Here, maybe we could just allocate this once during device/buffer
initialization
Quoted 1 line
and allocate it for the maximum possible size, i.e all scan elements
enabled...
Quoted 7 lines
We can but to do that requires more invasive changes. Perhaps not horrible ones,
but we'd need to add stuff to caller drivers to configure the bounce buffer, or
some flag to see they needed to enable support for this. Only 3 drivers out of
all those in IIO are currently using this code (might be a few othersthat
Quoted 2 lines
could
take advantage but not many - so need to keep it localized).I was not thinking that way. I was just thinking in always allocating the bounce
buffer. But I do agree that it is less than ideal to allocate something that will
almost never be used...Quoted 4 lines
I suppose we could have an
iio_unaligned_push_to_buffers_prepare()to be called from theupdate_scan_mode()
callback (typed to be useable as that callback if nothing else to bedone
Quoted 2 lines
in
the relevant drivers.It could also be an option... If we want to not use realloc, we would have to manually
control the lifespan of the buffer and free it in the disable path. But I guess that might feel
like too much handling...
Quoted 9 lines
Anyways, maybe sticking with realloc, we could just use a more normal pattern like:
tmp =
devm_realloc(bounce_fuffer, ...);
if (!tmp)error;
bounce_buffer= tmp;
I just read your reply to Andy about this and I think you are right... If the new allocation fails, the older pointer is still preserved. However, I find the above form a bit more readable and straight :)
- Nuno Sá
Re: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
Quoted 2 lines · Nuno
Quoted 2 lines
From: Sa, Nuno <[email protected]>
Sent: Monday, May 3, 2021 9:15 AM
...
Quoted 7 lines
Quoted 7 lines
Anyways, maybe sticking with realloc, we could just use a more normal pattern like:
tmp =
devm_realloc(bounce_fuffer, ...);
if (!tmp)error;
Quoted 6 lines
Quoted 2 lines
bounce_buffer= tmp;I just read your reply to Andy about this and I think you are right... If the new allocation fails, the older pointer is still preserved. However, I find the above form a bit more readable and straight :)
You beat me up to it.
While devm_realloc() may actually be safer, the pattern is that: we should not rely on specifics of the custom implementation, it will really confuse new contributors or people who may take this as a "good' (which is not) pattern.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Jonathan Cameron
maintainerRe: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
In reply to Sa, NunoQuoted 118 lines · Nuno
Quoted 118 lines
-----Original Message-----
From: Sa, Nuno <[email protected]>
Sent: Monday, May 3, 2021 9:15 AM
To: Jonathan Cameron <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: RE: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 22 lines
-----Original Message-----
From: Jonathan Cameron <[email protected]>
Sent: Sunday, May 2, 2021 6:09 PM
To: Sa, Nuno <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: Re: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 9 lines · Nuno
Quoted 9 lines
From: Jonathan Cameron <[email protected]>
Sent: Saturday, May 1, 2021 7:25 PM
To: [email protected]
Cc: Jonathan Cameron <[email protected]>
Subject: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.From: Jonathan Cameron <[email protected]>
Quoted 9 lines
Quoted 3 lines
Quoted 3 lines
Whilst it is almost always possible to arrange for scan data to be read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a few placeswhere
Quoted 3 lines
Quoted 3 lines
leading data needs to be skipped over.
For these cases introduce a function that will allocate an
appropriate
Quoted 1 line
Quoted 1 line
sized and aligned bounce buffer (if not already allocated) and
copy
Quoted 25 lines
Quoted 6 lines
Quoted 6 lines
the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bounce buffer. We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.Signed-off-by Jonathan CameronQuoted 8 lines
Review
Quoted 8 lines
--- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 3 files changed, 44 insertions(+)(quoted diff) +0
-0b/drivers/iio/industrialio-
Quoted 9 lines
Quoted 9 lines
buffer.c
index9a8e16c7e9af..818dfaa73665100644drivers/iio/industrialio-buffer.c +0
-0@@ -1727,6 +1727,42 @@ int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) }EXPORT_SYMBOL_GPL (iio_push_to_buffers);Quoted 2 lines
Quoted 2 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1732
Quoted 2 lines
drivers/iio/industrialio-buffer.c +2
-0+/** +* iio_push_to_buffers_with_ts_na() - push to registered buffer,
Quoted 11 lines
Quoted 7 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data)
Quoted 7 lines
drivers/iio/industrialio-buffer.c +7
-0+* no alignment or space requirements. +* @indio_dev: iio_dev structure for device. +* @data: channel data excluding the timestamp. +* @data_sz: size of data. +* @timestamp: timestamp for the sample data. +* +* This special variant of iio_push_to_buffers_with_timetamp() does
Quoted 1 line
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1740
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+* not require space for the timestamp, or 8 byte alignment of data.
Quoted 1 line
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+* It does however require an allocation on first call and additional
Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
drivers/iio/industrialio-buffer.c +4
-0+* coppies on all calls, so should be avoided if possible +*/ +int iio_push_to_buffers_with_ts_na(struct iio_dev *indio_dev, + const void *data,Quoted 8 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data) · line 1750
Quoted 7 lines
drivers/iio/industrialio-buffer.c +6
-0+ size_t data_sz, + int64_t timestamp) +{ +struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz =min_t (size_t, indio_dev->scan_bytes, data_sz);I'm not really sure about this. Is it really a good idea to silently
Quoted 4 lines
truncate
Quoted 1 line
the data if some erroneous driver calls this with
data_sz>
scan_bytes?Quoted 1 line
(I don't think it's ever valid for
data_szto be bigger than
scan_bytes)Quoted 5 lines
It's not probably not. This is just a fairly nasty way of papering over anyone doing that.
Quoted 1 line
We might be discarding data that userland actually requested. So,
I'm
Quoted 10 lines
Quoted 1 line
not sure if it would not be better to be strict here and return error,
so that
Quoted 1 line
driver developers could catch this early on...
I'm fairly sure we'll never be discarding data userspace requested because
scan_byteswill be the maximum possible and is the size of the
kfifo element. Any more data that comes in is going to be dropped anyway
assuming it ends up in akfifo_put()
Quoted 15 lines
Quoted 7 lines
I think it depends. Think on a totally erroneous driver that does not track
enabled elements and just pushes all elements to IIO buffers. If userspace
just requested, let's say, the last 2 elements, I believe those would be definitely discarded by this... More subtle things could also happen I guess...Quoted 8 lines
In this case, we can arguably say that the driver is completely broken and while
I'm fairly sure such a scenario would never pass through revision, I think that
if we make things more strict, we might minimize the chances of such patches
to appear in the first place (unless someone is sending patches without testing them at all).
Like me ;) *laughs*
Agreed there is nothing stopping at least adding a warning print here and it might help weed out gross bugs even if it is imprecise.
Quoted 25 lines
Quoted 25 lines
Quoted 5 lines
I wondered if we wanted to enforce this? It would not seem totally silly for a driver that wasn't bothering to track the size of it's enabled channels to provide
data_sz== maximum size it ever needs. Docs should definitely say if that's valid however, and it's definitelyless
Quoted 6 lines
than optimal because we may copy garbage that will then be overwritten by
the timestamp.Quoted 2 lines
data_sz<scan_bytesis also probably not valid as that mighthappen
Quoted 2 lines
Quoted 2 lines
if some drivers call this without the scan elements properly aligned which will definitely be problematic... Naturally in this case we
would
Quoted 1 line
Quoted 1 line
have to take into account timestamp possible alignment fill bytes as
it
Quoted 1 line
Quoted 1 line
could be possible (and likely) that
data_szwon't include thosebits...
Quoted 3 lines
data_sz<scan_bytesis intentionally valid (and indeed I made one of the drivers do this). Given we are bouncing anyway, why insistthe
Quoted 1 line
driver provides space for padding + timestamp? That can be done
Quoted 8 lines
Quoted 8 lines
for
Quoted 1 line
the bounce buffer.
What I meant was that if
data_sz<scan_bytes- timestamp - padding, then
the driver is most likely not guaranteeing properly aligned elements. I was
not that clear :)
If you hit that condition I think it would be a bug as not enough data provided. We could easily enforce data_sz < scan_bytes - timstamp, but the padding would need to be stashed somewhere. + I've been lazy in one of the drivers because I couldn't figure it out in a trivial fashion - perhaps I should take another look at that one and see if I can make the check accurate rather than a worst case bound.
Quoted 49 lines
Quoted 49 lines
Anyways, I still feel that if we are taking the
data_szparameter on the API,
we should handle it more strictly. But I do not feel too strong about this (as
you said, It's not expected to exist too much users of this) and in the end
of the day it still behaves the same way as 'iio_push_to_buffers_with_timestamp()'...Quoted 13 lines
Quoted 2 lines
Alternatively, we could be consistent with
'
iio_push_to_buffers_with_timestamp()'Quoted 4 lines
and assume drivers play along in terms of alignment and size (so, removing the
data_szparameter)...My point is, if we are taking the
data_szparameter, maybe weshould be pedantic
Quoted 5 lines
about it...
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev · line 1754
Quoted 2 lines
drivers/iio/industrialio-buffer.c +1
-0+if (iio_dev_opaque->bounce_buffer_size != indio_dev- >scan_bytes) {Where do we set '
bounce_buffer_size'? I guess we do not wantto
Quoted 17 lines
Quoted 1 line
get here all the time...
:( oops
Quoted 7 lines
Quoted 5 lines
drivers/iio/industrialio-buffer.c +4
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque->bounce_buffer, + indio_dev->scan_bytes,
GFP_KERNEL);I'm also not a big fan of realloc so I prefer to avoid it unless really
needed...
Quoted 1 line
Here, maybe we could just allocate this once during device/buffer
initialization
Quoted 1 line
and allocate it for the maximum possible size, i.e all scan elements
enabled...
Quoted 7 lines
We can but to do that requires more invasive changes. Perhaps not horrible ones,
but we'd need to add stuff to caller drivers to configure the bounce buffer, or
some flag to see they needed to enable support for this. Only 3 drivers out of
all those in IIO are currently using this code (might be a few othersthat
Quoted 8 lines
Quoted 8 lines
Quoted 2 lines
could
take advantage but not many - so need to keep it localized).I was not thinking that way. I was just thinking in always allocating the bounce
buffer. But I do agree that it is less than ideal to allocate something that will
almost never be used...
Ah got you. We could do that, but I think it wants to be opt in just because I'd feel bad adding an extra allocation to every driver.
Quoted 22 lines
Quoted 22 lines
Quoted 5 lines
I suppose we could have an
iio_unaligned_push_to_buffers_prepare()to be called from theupdate_scan_mode()
callback (typed to be useable as that callback if nothing else to bedone
Quoted 2 lines
in
the relevant drivers.It could also be an option... If we want to not use realloc, we would have to manually
control the lifespan of the buffer and free it in the disable path. But I guess that might feel
like too much handling...Anyways, maybe sticking with realloc, we could just use a more normal pattern like:
tmp =
devm_realloc(bounce_fuffer, ...);
if (!tmp)error;
Quoted 7 lines
Quoted 2 lines
bounce_buffer= tmp;I just read your reply to Andy about this and I think you are right... If the new allocation fails, the older pointer is still preserved. However, I find the above form a bit more readable and straight :)
Understood. Given it's causing confusion (and I had to think about it in the first place), simply have the more standard pattern is probably wise even if strictly speaking unnecessary! Give people what they expect to see rather than being 'clever' :)
Quoted 1 line
- Nuno Sá
Sa, Nuno
maintainerRE: [RFC PATCH 1/4] iio: core: Introduce iio_push_to_buffers_with_ts_na() for non aligned case.
Quoted 161 lines
-----Original Message-----
From: Jonathan Cameron <[email protected]>
Sent: Monday, May 3, 2021 12:40 PM
To: Sa, Nuno <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: Re: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 32 lines · Nuno
Quoted 32 lines
-----Original Message-----
From: Sa, Nuno <[email protected]>
Sent: Monday, May 3, 2021 9:15 AM
To: Jonathan Cameron <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: RE: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 20 lines
-----Original Message-----
From: Jonathan Cameron <[email protected]>
Sent: Sunday, May 2, 2021 6:09 PM
To: Sa, Nuno <[email protected]>
Cc: [email protected]; Jonathan Cameron <[email protected]>
Subject: Re: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 7 lines · Nuno
Quoted 7 lines
From: Jonathan Cameron <[email protected]>
Sent: Saturday, May 1, 2021 7:25 PM
To: [email protected]
Cc: Jonathan Cameron <[email protected]>
Subject: [RFC PATCH 1/4] iio: core: Introduceiio_push_to_buffers_with_ts_na()for non aligned case.Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
From: Jonathan Cameron <[email protected]>
Whilst it is almost always possible to arrange for scan data to
be
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
read directly into a buffer that is suitable for passing to
iio_push_to_buffers_with_timestamp(), there are a fewplaces
Quoted 9 lines
Quoted 9 lines
Quoted 6 lines
where
Quoted 3 lines
Quoted 3 lines
leading data needs to be skipped over.
For these cases introduce a function that will allocate an
appropriate
Quoted 1 line
Quoted 1 line
sized and aligned bounce buffer (if not already allocated) and
copy
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
the unaligned data into that before calling
iio_push_to_buffers_with_timestamp()on the bouncebuffer.
Quoted 12 lines
Quoted 12 lines
Quoted 12 lines
Quoted 4 lines
Quoted 4 lines
We tie the lifespace of this buffer to that of the iio_dev.dev which should ensure no memory leaks occur.
Signed-off-by Jonathan CameronQuoted 7 lines
Quoted 7 lines
--- drivers/iio/industrialio-buffer.c | 36 +++++++++++++++++++++++++++++++ include/linux/iio/buffer.h | 4 ++++ include/linux/iio/iio-opaque.h | 4 ++++ 3 files changed, 44 insertions(+)Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 1 line
Review
Quoted 1 line
(quoted diff) +0
-0b/drivers/iio/industrialio-
Quoted 5 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct
Quoted 5 lines
buffer.c
index9a8e16c7e9af..818dfaa73665100644drivers/iio/industrialio-buffer.c +0
-0@@ -1727,6 +1727,42 @@ int iio_push_to_buffers(struct
iio_devQuoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
drivers/iio/industrialio-buffer.c +2
-0*indio_dev, const void *data) }EXPORT_SYMBOL_GPL (iio_push_to_buffers); +/** +* iio_push_to_buffers_with_ts_na() - push to registered Quoted 8 lines
Quoted 8 lines
buffer,
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
drivers/iio/industrialio-buffer.c +7
-0+* no alignment or space requirements. +* @indio_dev: iio_dev structure for device. +* @data: channel data excluding the timestamp. +* @data_sz: size of data. +* @timestamp: timestamp for the sample data. +* +* This special variant of
iio_push_to_buffers_with_timetamp()Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
does
Quoted 1 line
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+* not require space for the timestamp, or 8 byte alignment of
Quoted 6 lines
Quoted 6 lines
Quoted 2 lines
data.
Quoted 1 line
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+* It does however require an allocation on first call and additional
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
drivers/iio/industrialio-buffer.c +3
-0+* coppies on all calls, so should be avoided if possible +*/ +int iio_push_to_buffers_with_ts_na(struct iio_dev
*indio_dev,Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data)
Quoted 8 lines
drivers/iio/industrialio-buffer.c +7
-0+const void *data, + size_t data_sz, + int64_t timestamp) +{ + struct iio_dev_opaque *iio_dev_opaque =to_iio_dev_opaque (indio_dev); + + data_sz = min_t(size_t, indio_dev->scan_bytes,
data_sz);Quoted 8 lines
Quoted 8 lines
Quoted 5 lines
Quoted 1 line
I'm not really sure about this. Is it really a good idea to silently
truncate
Quoted 1 line
the data if some erroneous driver calls this with
data_sz>
scan_bytes?Quoted 1 line
(I don't think it's ever valid for
data_szto be bigger than
scan_bytes)Quoted 2 lines
It's not probably not. This is just a fairly nasty way of papering
over
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
anyone doing that.
Quoted 1 line
We might be discarding data that userland actually requested.
So,
Quoted 2 lines
Quoted 2 lines
I'm
Quoted 1 line
Quoted 1 line
not sure if it would not be better to be strict here and return
error,
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
so that
Quoted 1 line
driver developers could catch this early on...
I'm fairly sure we'll never be discarding data userspace requested because
scan_byteswill be the maximum possible and is the sizeof
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
the
kfifo element. Any more data that comes in is going to bedropped
Quoted 8 lines
Quoted 8 lines
Quoted 2 lines
anyway
assuming it ends up in akfifo_put()I think it depends. Think on a totally erroneous driver that does not track
enabled elements and just pushes all elements to IIO buffers. If userspace
just requested, let's say, the last 2 elements, I believe those wouldbe
Quoted 1 line
Quoted 1 line
definitely discarded by this... More subtle things could also happen
I
Quoted 2 lines
Quoted 2 lines
guess...
In this case, we can arguably say that the driver is completelybroken
Quoted 4 lines
Quoted 4 lines
and while
I'm fairly sure such a scenario would never pass through revision, I think that
if we make things more strict, we might minimize the chances of
Quoted 7 lines
such
Quoted 2 lines
Quoted 2 lines
patches
to appear in the first place (unless someone is sending patcheswithout
Quoted 1 line
Quoted 1 line
testing them at all).
Like me ;) *laughs*
Well, at least you warned everyone about it :)
Quoted 46 lines
Agreed there is nothing stopping at least adding a warning print here and it might help weed out gross bugs even if it is imprecise.
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
I wondered if we wanted to enforce this? It would not seem
totally
Quoted 1 line
Quoted 1 line
Quoted 1 line
silly for a driver that wasn't bothering to track the size of it's
enabled
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
channels to provide
data_sz== maximum size it ever needs. Docs should definitely say if that's valid however, and it'sdefinitely
Quoted 9 lines
Quoted 9 lines
less
Quoted 6 lines
than optimal because we may copy garbage that will then be overwritten by
the timestamp.Quoted 2 lines
data_sz<scan_bytesis also probably not valid as that mighthappen
Quoted 1 line
Quoted 1 line
if some drivers call this without the scan elements properly
aligned
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
Quoted 1 line
which will definitely be problematic... Naturally in this case we
would
Quoted 1 line
Quoted 1 line
have to take into account timestamp possible alignment fill
bytes as
Quoted 5 lines
Quoted 5 lines
it
Quoted 1 line
Quoted 1 line
could be possible (and likely) that
data_szwon't include thosebits...
Quoted 2 lines
data_sz<scan_bytesis intentionally valid (and indeed I madeone
Quoted 3 lines
Quoted 3 lines
Quoted 1 line
of the drivers do this). Given we are bouncing anyway, why insist
the
Quoted 1 line
driver provides space for padding + timestamp? That can be
done
Quoted 4 lines
Quoted 4 lines
for
Quoted 1 line
the bounce buffer.
What I meant was that if
data_sz<scan_bytes- timestamp -padding,
Quoted 2 lines
Quoted 2 lines
then
the driver is most likely not guaranteeing properly alignedelements. I
Quoted 2 lines
Quoted 2 lines
was
not that clear :)
Quoted 6 lines
If you hit that condition I think it would be a bug as not enough data provided.
We could easily enforcedata_sz<scan_bytes- timstamp, but the padding would
need to be stashed somewhere. + I've been lazy in one of the drivers because
I think with data_sz we can easily calculate the needed padding for the timestamp... Naturally we would need to clearly state in the docs (probably done already) that data_sz should not contain any bits related with the timestamp element...
Quoted 70 lines
I couldn't figure it out in a trivial fashion - perhaps I should take another look at that one and see if I can make the check accurate rather than a worst
case bound.Quoted 2 lines
Quoted 2 lines
Anyways, I still feel that if we are taking the
data_szparameter onthe
Quoted 4 lines
Quoted 4 lines
API,
we should handle it more strictly. But I do not feel too strong about this (as
you said, It's not expected to exist too much users of this) and inthe
Quoted 6 lines
Quoted 6 lines
end
of the day it still behaves the same way as
'iio_push_to_buffers_with_timestamp()'...Quoted 2 lines
Quoted 2 lines
Alternatively, we could be consistent with
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
'
iio_push_to_buffers_with_timestamp()'Quoted 4 lines
and assume drivers play along in terms of alignment and size (so, removing the
data_szparameter)...My point is, if we are taking the
data_szparameter, maybe weshould be pedantic
Quoted 2 lines
about it...
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct · line 1753
Quoted 1 line
drivers/iio/industrialio-buffer.c +1
-0+ if (iio_dev_opaque->bounce_buffer_size !=indio_dev-
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct · line 1754
Quoted 1 line
drivers/iio/industrialio-buffer.c +0
-0>scan_bytes) {Where do we set '
bounce_buffer_size'? I guess we do notwant
Quoted 13 lines
Quoted 8 lines
to
Quoted 7 lines
Quoted 1 line
get here all the time...
:( oops
Quoted 3 lines
drivers/iio/industrialio-buffer.c int iio_push_to_buffers(struct iio_dev *indio_dev, const void *data)
Quoted 3 lines
drivers/iio/industrialio-buffer.c +3
-0+ iio_dev_opaque->bounce_buffer = + devm_krealloc(&indio_dev->dev, + iio_dev_opaque-
bounce_buffer,Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 2 lines
drivers/iio/industrialio-buffer.c +1
-0+ indio_dev->scan_bytes,
GFP_KERNEL);I'm also not a big fan of realloc so I prefer to avoid it unless
really
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
needed...
Quoted 1 line
Here, maybe we could just allocate this once during
device/buffer
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
initialization
Quoted 1 line
and allocate it for the maximum possible size, i.e all scan
elements
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
enabled...
We can but to do that requires more invasive changes. Perhaps
not
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
horrible ones,
but we'd need to add stuff to caller drivers to configure thebounce
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
buffer, or
some flag to see they needed to enable support for this. Only 3 drivers out of
all those in IIO are currently using this code (might be a fewothers
Quoted 5 lines
Quoted 5 lines
that
Quoted 2 lines
could
take advantage but not many - so need to keep it localized).I was not thinking that way. I was just thinking in always allocating
the
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
bounce
buffer. But I do agree that it is less than ideal to allocate something that will
almost never be used...Ah got you. We could do that, but I think it wants to be opt in just because
I'd feel bad adding an extra allocation to every driver.
Agreed...
- Nuno Sá
Re: [RFC PATCH 3/4] iio: gyro: mpu3050: Fix alignment and size issues with buffers.
In reply to Jonathan CameronQuoted 9 lines · Jonathan Cameron
From: Jonathan Cameron <[email protected]>
Fix a set of closely related issues.
1. When usingfifo_values()there was not enough space for the timestamp to be inserted byiio_push_to_buffers_with_timestamp()
2.fifo_values()did not meet the alignment requirement ofiio_push_to_buffers_with_timestamp()
3.hw_valuesdid not meet the alignment requirement either.
Quoted 8 lines
1 and 2 fixed by using new
iio_push_to_buffers_with_ts_na()which has no alignment or space padding requirements.
3 fixed by introducing a structure that makes the space and alignment requirements explicit.Fixes 3904b28efb2c ("iio: gyro: Add driver for the MPU-3050 gyroscope")Signed-off-by Jonathan Cameron <[email protected]>Cc Linus Walleij <[email protected]>
Again a very nice fix:
Again I wonder how we can help driver authors to always get this right.
Yours,
Linus Walleij