A-Team_Digital_Solutions/android_kernel_motorola_sm6375
Watch
2
0
Fork
You've already forked android_kernel_motorola_sm6375
 0
mirror of https://github.com/BobTheBlinker/android_kernel_motorola_sm6375.git synced 2026-10-12 07:37:31 -04:00
Code Issues Projects Releases Packages Wiki Activity

msm: mhi_dev: Ensure the flush list is empty for a stopped channel

Browse source 
If a channel gets stopped, while there are some outstanding requests/
transfers with IPA, then we simply ignore the completion from IPA.
But we must have already added the transfer-completion-event to the
flush list before submitting the transfer to IPA.

There is a possibility that we flush these stale events along with new
flush events (for new requests) when the channel gets re-started.
This un-intentional completion event can lead to out-of-sequence
events at the host.

So while stopping a channel, ensure all elements in the flush list
of that channel are discarded and the flush list is empty.

Change-Id: I04445b1f9f4d2d4fce0b8a96277517de14481ead
Signed-off-by: Veerabhadrarao Badiganti <vbadigan@codeaurora.org>
This commit is contained in:
Veerabhadrarao Badiganti 2021-09-06 21:39:13 +05:30 • committed by Gerrit - the friendly Code Review server
commit 4b3ca62138
1 changed files with 16 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

16
drivers/platform/msm/mhi_dev/mhi.c
View file

@ -1841,6 +1841,8 @@ static void mhi_dev_process_reset_cmd(struct mhi_dev *mhi, int ch_id)
int rc = 0;
struct mhi_dev_channel *ch;
struct mhi_addr host_addr;
struct event_req *itr, *tmp;
unsigned long flags;
rc = mhi_dev_mmio_disable_chdb_a7(mhi, ch_id);
if (rc) {
@ -1855,6 +1857,20 @@ static void mhi_dev_process_reset_cmd(struct mhi_dev *mhi, int ch_id)
}
ch = &mhi->ch[ch_id];
mhi_log(MHI_MSG_VERBOSE, "Processing reset cmd for ch%d\n", ch_id);
/*
* Ensure that the completions that are present in the flush list are
* removed from the list and discarded before stopping the channel.
* Otherwise, those stale events may get flushed along with a valid
* event in the next flush operation.
*/
spin_lock_irqsave(&mhi_ctx->lock, flags);
list_for_each_entry_safe(itr, tmp, &ch->flush_event_req_buffers, list) {
list_del(&itr->list);
kfree(itr);
}
spin_unlock_irqrestore(&mhi_ctx->lock, flags);
/* hard stop and set the channel to stop */
mhi->ch_ctx_cache[ch_id].ch_state =
MHI_DEV_CH_STATE_DISABLED;