On 19 August 2011 16:28, Per Forlin <per.forlin@xxxxxxxxxxxxxx> wrote:
> From: Per Forlin <per.forlin@xxxxxxxxxx>
>
> FSG_NUM_BUFFERS is set to 2 as default.
> Usually 2 buffers are enough to establish a good buffering pipeline.
> The number may be increased in order to compensate a for bursty VFS
> behaviour.
>
> Here follows a description of system that may require more than
> 2 buffers.
> * CPU ondemand governor active
> * latency cost for wake up and/or frequency change
> * DMA for IO
>
> Use case description.
> * Data transfer from MMC via VFS to USB.
> * DMA shuffles data from MMC and to USB.
> * The CPU wakes up every now and then to pass data in and out from VFS,
> which cause the bursty VFS behaviour.
>
> Test set up
> * Running dd on the host reading from the mass storage device
> * cmdline: dd if=/dev/sdb of=/dev/null bs=4k count=$((256*100))
> * Caches are dropped on the host and on the device before each run
>
> Measurements on a Snowball board with ondemand_govenor active.
>
> FSG_NUM_BUFFERS 2
> 104857600 bytes (105 MB) copied, 5.62173 s, 18.7 MB/s
> 104857600 bytes (105 MB) copied, 5.61811 s, 18.7 MB/s
> 104857600 bytes (105 MB) copied, 5.57817 s, 18.8 MB/s
>
> FSG_NUM_BUFFERS 4
> 104857600 bytes (105 MB) copied, 5.26839 s, 19.9 MB/s
> 104857600 bytes (105 MB) copied, 5.2691 s, 19.9 MB/s
> 104857600 bytes (105 MB) copied, 5.2711 s, 19.9 MB/s
>
> There may not be one optimal number for all boards. This is why
> the number is added to Kconfig. If selecting USB_GADGET_DEBUG_FILES
> this value may be set by a module parameter as well.
>
> Signed-off-by: Per Forlin <per.forlin@xxxxxxxxxx>
> ---
> Change log.
> v2: Update after proofreading comments from Michal Nazarewicz
> v3: Clarify the description of this patch based on input from Alan Stern
> v4: - Introduce a module_param to set number of pipeline buffers
> if USB_DEBUG is set. In order to add this support fsg_common is
> allocated at runtime. The fsg_buffhd list size is appended to fsg_dev
> and fsg_common at runtime allocation.
> - The previous acks from Michal and Alan on v3 are not applicable
> for this version since it's a new implementation.
> v5: - Separate allocation of buffhds in f_mass_storage.c
> - Change FSG_NUM_BUFFERS to fsg_num_buffers to indicate variable value.
> - code clean up and document module param, all thanks to Michal and Alan.
>
> drivers/usb/gadget/Kconfig | 16 ++++++++++++++++
> drivers/usb/gadget/f_mass_storage.c | 24 ++++++++++++++++--------
> drivers/usb/gadget/file_storage.c | 26 ++++++++++++++++----------
> drivers/usb/gadget/mass_storage.c | 3 +++
> drivers/usb/gadget/multi.c | 3 +++
> drivers/usb/gadget/storage_common.c | 25 +++++++++++++++++++++++--
> 6 files changed, 77 insertions(+), 20 deletions(-)
>
> diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig
> index 44b6b40..65ceecf 100644
> --- a/drivers/usb/gadget/Kconfig
> +++ b/drivers/usb/gadget/Kconfig
> @@ -96,6 +96,22 @@ config USB_GADGET_VBUS_DRAW
> This value will be used except for system-specific gadget
> drivers that have more specific information.
>
> +config USB_GADGET_STORAGE_NUM_BUFFERS
> + int "Number of storage pipeline buffers"
> + range 2 4
> + default 2
> + help
> + Usually 2 buffers are enough to establish a good buffering
> + pipeline. The number may be increased in order to compensate
> + for a bursty VFS behaviour. For instance there may be cpu wake up
> + latencies that makes the VFS to appear bursty in a system with
> + an cpu on-demand governor. Especially if DMA is doing IO to
> + offload the CPU. In this case the CPU will go into power
> + save often and spin up occasionally to move data within VFS.
> + If selecting USB_GADGET_DEBUG_FILES this value may be set by
> + a module parameter as well.
> + If unsure, say 2.
> +
> #
> # USB Peripheral Controller Support
> #
> diff --git a/drivers/usb/gadget/f_mass_storage.c b/drivers/usb/gadget/f_mass_storage.c
> index 5b93395..49da6a1 100644
> --- a/drivers/usb/gadget/f_mass_storage.c
> +++ b/drivers/usb/gadget/f_mass_storage.c
> @@ -363,7 +363,7 @@ struct fsg_common {
>
> struct fsg_buffhd *next_buffhd_to_fill;
> struct fsg_buffhd *next_buffhd_to_drain;
> - struct fsg_buffhd buffhds[FSG_NUM_BUFFERS];
> + struct fsg_buffhd *buffhds;
>
> int cmnd_size;
> u8 cmnd[MAX_COMMAND_SIZE];
> @@ -2348,7 +2348,7 @@ reset:
> if (common->fsg) {
> fsg = common->fsg;
>
> - for (i = 0; i < FSG_NUM_BUFFERS; ++i) {
> + for (i = 0; i < fsg_num_buffers; ++i) {
> struct fsg_buffhd *bh = &common->buffhds[i];
>
> if (bh->inreq) {
> @@ -2406,7 +2406,7 @@ reset:
> clear_bit(IGNORE_BULK_OUT, &fsg->atomic_bitflags);
>
> /* Allocate the requests */
> - for (i = 0; i < FSG_NUM_BUFFERS; ++i) {
> + for (i = 0; i < fsg_num_buffers; ++i) {
> struct fsg_buffhd *bh = &common->buffhds[i];
>
> rc = alloc_request(common, fsg->bulk_in, &bh->inreq);
> @@ -2475,7 +2475,7 @@ static void handle_exception(struct fsg_common *common)
>
> /* Cancel all the pending transfers */
> if (likely(common->fsg)) {
> - for (i = 0; i < FSG_NUM_BUFFERS; ++i) {
> + for (i = 0; i < fsg_num_buffers; ++i) {
> bh = &common->buffhds[i];
> if (bh->inreq_busy)
> usb_ep_dequeue(common->fsg->bulk_in, bh->inreq);
> @@ -2487,7 +2487,7 @@ static void handle_exception(struct fsg_common *common)
> /* Wait until everything is idle */
> for (;;) {
> int num_active = 0;
> - for (i = 0; i < FSG_NUM_BUFFERS; ++i) {
> + for (i = 0; i < fsg_num_buffers; ++i) {
> bh = &common->buffhds[i];
> num_active += bh->inreq_busy + bh->outreq_busy;
> }
> @@ -2510,7 +2510,7 @@ static void handle_exception(struct fsg_common *common)
> */
> spin_lock_irq(&common->lock);
>
> - for (i = 0; i < FSG_NUM_BUFFERS; ++i) {
> + for (i = 0; i < fsg_num_buffers; ++i) {
> bh = &common->buffhds[i];
> bh->state = BUF_STATE_EMPTY;
> }
> @@ -2737,6 +2737,14 @@ static struct fsg_common *fsg_common_init(struct fsg_common *common,
> common->free_storage_on_release = 0;
> }
>
> + common->buffhds = kcalloc(fsg_num_buffers,
> + sizeof *(common->buffhds), GFP_KERNEL);
Need to add a kfree() at release too. This data is never freed
otherwise. I'll update fix that in the next version together with any
new comments from Michal and Alan.
/Per
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
