lkml.org 
[lkml]   [2021]   [Aug]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH v2] staging: greybus: Convert uart.c from IDR to XArray
From
Date
On 8/14/21 1:11 PM, Fabio M. De Francesco wrote:
> Convert greybus/uart.c from IDR to XArray. The abstract data type XArray
> is more memory-efficient, parallelisable, and cache friendly. It takes
> advantage of RCU to perform lookups without locking. Furthermore, IDR is
> deprecated because XArray has a better (cleaner and more consistent) API.

I haven't verified the use of the new API (yet) but I have a few
comments on your patch, below.

-Alex

> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>

I'm not sure I'm right about this... But the actual change you're
making has nothing to do with what the Intel test robot reported.
I personally find the "Reported-by" here a little misleading, but
maybe the "Link" line that gets added will provide explanation.

Anyway, unless someone else contradicts/corrects me, I'd rather
not have the "Reported-by" here (despite wanting to provide much
credit to <lkp@intel.com>...).

> ---
>
> v1->v2:
> Fixed an issue found by the kernel test robot. It was due to
> passing to xa_*lock() the same old mutex that IDR used with
> the previous version of the code.
>
> drivers/staging/greybus/uart.c | 29 ++++++++++++++---------------
> 1 file changed, 14 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/staging/greybus/uart.c b/drivers/staging/greybus/uart.c
> index 73f01ed1e5b7..5bf993e40f84 100644
> --- a/drivers/staging/greybus/uart.c
> +++ b/drivers/staging/greybus/uart.c
> @@ -22,7 +22,7 @@
> #include <linux/serial.h>
> #include <linux/tty_driver.h>
> #include <linux/tty_flip.h>
> -#include <linux/idr.h>
> +#include <linux/xarray.h>
> #include <linux/fs.h>
> #include <linux/kdev_t.h>
> #include <linux/kfifo.h>
> @@ -33,6 +33,7 @@
> #include "gbphy.h"
>
> #define GB_NUM_MINORS 16 /* 16 is more than enough */
> +#define GB_RANGE_MINORS XA_LIMIT(0, GB_NUM_MINORS)
> #define GB_NAME "ttyGB"

Please align the right-hand side of all three definitions here.

> #define GB_UART_WRITE_FIFO_SIZE PAGE_SIZE
> @@ -67,8 +68,7 @@ struct gb_tty {
> };
>
> static struct tty_driver *gb_tty_driver;
> -static DEFINE_IDR(tty_minors);
> -static DEFINE_MUTEX(table_lock);
> +static DEFINE_XARRAY(tty_minors);
>
> static int gb_uart_receive_data_handler(struct gb_operation *op)
> {
> @@ -77,6 +77,7 @@ static int gb_uart_receive_data_handler(struct gb_operation *op)
> struct tty_port *port = &gb_tty->port;
> struct gb_message *request = op->request;
> struct gb_uart_recv_data_request *receive_data;
> +

Please do not add a blank line amid the local variable
definitions.

I'm not sure it checks for this, but you should run
your patch through "checkpatch.pl" before you send
it. E.g.:
./scripts/checkpatch.pl idr_to_xarray.patch

The error reported in the build of your first version
of this patch makes me think you might not have test-
built the code. I don't know if that's the case, but
(at least) building the code is expected before you
submit a patch for review.

> u16 recv_data_size;
> int count;
> unsigned long tty_flags = TTY_NORMAL;
> @@ -341,8 +342,8 @@ static struct gb_tty *get_gb_by_minor(unsigned int minor)
> {
> struct gb_tty *gb_tty;
>
> - mutex_lock(&table_lock);
> - gb_tty = idr_find(&tty_minors, minor);
> + xa_lock(&tty_minors);
> + gb_tty = xa_load(&tty_minors, minor);
> if (gb_tty) {
> mutex_lock(&gb_tty->mutex);
> if (gb_tty->disconnected) {
> @@ -353,19 +354,19 @@ static struct gb_tty *get_gb_by_minor(unsigned int minor)
> mutex_unlock(&gb_tty->mutex);
> }
> }
> - mutex_unlock(&table_lock);
> + xa_unlock(&tty_minors);
> return gb_tty;
> }
>
> static int alloc_minor(struct gb_tty *gb_tty)
> {
> int minor;
> + int ret;
>
> - mutex_lock(&table_lock);
> - minor = idr_alloc(&tty_minors, gb_tty, 0, GB_NUM_MINORS, GFP_KERNEL);
> - mutex_unlock(&table_lock);
> - if (minor >= 0)
> - gb_tty->minor = minor;
> + ret = xa_alloc(&tty_minors, &minor, gb_tty, GB_RANGE_MINORS, GFP_KERNEL);
> + if (ret)
> + return ret;

The caller of alloc_minor() (gb_uart_probe()) checks the return
value, and if it's -ENOSPC it logs a device error indicating
there are no remaining free device minor numbers. For xa_alloc()
this is indicated by returning -EBUSY. Please update the caller
to print the error message based on the updated error code.

> + gb_tty->minor = minor;
> return minor;
> }
>
> @@ -374,9 +375,7 @@ static void release_minor(struct gb_tty *gb_tty)
> int minor = gb_tty->minor;
>
> gb_tty->minor = 0; /* Maybe should use an invalid value instead */
> - mutex_lock(&table_lock);
> - idr_remove(&tty_minors, minor);
> - mutex_unlock(&table_lock);
> + xa_erase(&tty_minors, minor);
> }
>
> static int gb_tty_install(struct tty_driver *driver, struct tty_struct *tty)
> @@ -982,7 +981,7 @@ static void gb_tty_exit(void)
> {
> tty_unregister_driver(gb_tty_driver);
> put_tty_driver(gb_tty_driver);
> - idr_destroy(&tty_minors);
> + xa_destroy(&tty_minors);
> }
>
> static const struct gbphy_device_id gb_uart_id_table[] = {
>

\
 
 \ /
  Last update: 2021-08-16 16:46    [W:0.105 / U:0.052 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site