Hi, On Fri, Jun 26, 2020 at 2:37 PM Doug Anderson <dianders@xxxxxxxxxxxx> wrote: > > Hi, > > On Fri, Jun 26, 2020 at 10:53 AM Rakesh Pillai <pillair@xxxxxxxxxxxxxx> wrote: > > > > Currently the NAPI processing loops through all > > the copy engines and processes a particular copy > > engine is the copy completion is set for that copy > > engine. The host driver is not supposed to access > > any copy engine register after clearing the interrupt > > status register. > > > > This might result in kernel crash like the one below > > [ 1159.220143] Call trace: > > [ 1159.220170] ath10k_snoc_read32+0x20/0x40 [ath10k_snoc] > > [ 1159.220193] ath10k_ce_per_engine_service_any+0x78/0x130 [ath10k_core] > > [ 1159.220203] ath10k_snoc_napi_poll+0x38/0x8c [ath10k_snoc] > > [ 1159.220270] net_rx_action+0x100/0x3b0 > > [ 1159.220312] __do_softirq+0x164/0x30c > > [ 1159.220345] run_ksoftirqd+0x2c/0x64 > > [ 1159.220380] smpboot_thread_fn+0x1b0/0x288 > > [ 1159.220405] kthread+0x11c/0x12c > > [ 1159.220423] ret_from_fork+0x10/0x18 > > > > To avoid such a scenario, we generate an interrupt > > summary by reading the copy completion for all the > > copy engine before actually processing any of them. > > This will avoid reading the interrupt status register > > for any CE after the interrupt status is cleared. > > > > Tested-on: WCN3990 hw1.0 SNOC WLAN.HL.3.1-01040-QCAHLSWMTPLZ-1 > > > > Signed-off-by: Rakesh Pillai <pillair@xxxxxxxxxxxxxx> > > --- > > drivers/net/wireless/ath/ath10k/ce.c | 63 ++++++++++++++++++++++-------------- > > drivers/net/wireless/ath/ath10k/ce.h | 5 +-- > > 2 files changed, 42 insertions(+), 26 deletions(-) > > I'm not an expert on this driver, but your change seems sane to me. > > Reviewed-by: Douglas Anderson <dianders@xxxxxxxxxxxx> > > With your patch I can no longer find a place to put in a magic delay > and reproduce the crash, thus: > > Tested-by: Douglas Anderson <dianders@xxxxxxxxxxxx> > > > If it matters, my WiFi firmware reports this: > > WLAN.HL.3.2.2-00490-QCAHLSWMTPL-1 > > ...and it should also be WCN3990. I should also note that, while I'm not terribly familiar with Kalle's workflow, I would have expected to see him in the "To:" list. I've added him, but it's possible he'll need you to repost the patch with him in the "To:" list. -Doug
