linux-kernel - Re: LKFT: arm x15: mmc1: cache flush error -110

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <f960aa98-5508-36fd-166d-7f41c7d85154@nvidia.com>
Date:   Tue, 25 Feb 2020 11:41:31 +0000
From:   Jon Hunter <jonathanh@...dia.com>
To:     Ulf Hansson <ulf.hansson@...aro.org>,
        Bitan Biswas <bbiswas@...dia.com>,
        Adrian Hunter <adrian.hunter@...el.com>
CC:     Naresh Kamboju <naresh.kamboju@...aro.org>,
        Jens Axboe <axboe@...nel.dk>,
        Alexei Starovoitov <ast@...nel.org>,
        linux-block <linux-block@...r.kernel.org>,
        <lkft-triage@...ts.linaro.org>,
        open list <linux-kernel@...r.kernel.org>,
        "linux-mmc@...r.kernel.org" <linux-mmc@...r.kernel.org>,
        Arnd Bergmann <arnd@...db.de>,
        John Stultz <john.stultz@...aro.org>,
        Thierry Reding <treding@...dia.com>
Subject: Re: LKFT: arm x15: mmc1: cache flush error -110


On 25/02/2020 10:04, Jon Hunter wrote:

...

>>>   I find that from the commit the changes in mmc_flush_cache below is
>>> the cause.
>>>
>>> ##
>>> @@ -961,7 +963,8 @@ int mmc_flush_cache(struct mmc_card *card)
>>>                          (card->ext_csd.cache_size > 0) &&
>>>                          (card->ext_csd.cache_ctrl & 1)) {
>>>                  err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
>>> -                               EXT_CSD_FLUSH_CACHE, 1, 0);
>>> +                                EXT_CSD_FLUSH_CACHE, 1,
>>> +                                MMC_CACHE_FLUSH_TIMEOUT_MS);
> 
> 
> I no longer see the issue on reverting the above hunk as Bitan suggested
> but now I see the following (which is expected) ...
> 
>  WARNING KERN mmc1: unspecified timeout for CMD6 - use generic

For Tegra, the default timeout used when no timeout is specified for CMD6
is 100mS. So hard-coding the following also appears to workaround the
problem on Tegra ...

diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
index 868653bc1555..5155e0240fca 100644
--- a/drivers/mmc/core/mmc_ops.c
+++ b/drivers/mmc/core/mmc_ops.c
@@ -992,7 +992,7 @@ int mmc_flush_cache(struct mmc_card *card)
                        (card->ext_csd.cache_size > 0) &&
                        (card->ext_csd.cache_ctrl & 1)) {
                err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
-                                EXT_CSD_FLUSH_CACHE, 1, 0);
+                                EXT_CSD_FLUSH_CACHE, 1, 100);
                if (err)
                        pr_err("%s: cache flush error %d\n",
                                        mmc_hostname(card->host), err);

So the problem appears to be causing by the timeout being too long rather
than not long enough. 

Looking more at the code, I think now that we are hitting the condition
...

diff --git a/drivers/mmc/core/mmc_ops.c b/drivers/mmc/core/mmc_ops.c
index 868653bc1555..feae82b1ff35 100644
--- a/drivers/mmc/core/mmc_ops.c
+++ b/drivers/mmc/core/mmc_ops.c
@@ -579,8 +579,10 @@ int __mmc_switch(struct mmc_card *card, u8 set, u8 index, u8 value,
         * the host to avoid HW busy detection, by converting to a R1 response
         * instead of a R1B.
         */
-       if (host->max_busy_timeout && (timeout_ms > host->max_busy_timeout))
+       if (host->max_busy_timeout && (timeout_ms > host->max_busy_timeout)) {
+               pr_warn("%s: timeout (%d) > max busy timeout (%d)", mmc_hostname(host), timeout_ms, host->max_busy_timeout);
                use_r1b_resp = false;
+       }


With the above I see ...

 WARNING KERN mmc1: timeout (1600) > max busy timeout (672)

So with the longer timeout we are not using/requesting the response.

Cheers
Jon

-- 
nvpublic