RE: [PATCH v2] mmc: inline the first mmc_scan() on mmc_start_host()
From: Biju Das
Date: Tue Jun 27 2023 - 14:10:02 EST
Hi Dennis,
Reverting commit 2cc83bf7d41113d9 ("mmc:core: Allow mmc_start_host() synchronously detect a card") fixes below issue[1] on RZ/G2L SMARC EVK too.
[1] Before reverting:
------------------
[ 2.469266] renesas_sdhi_internal_dmac 11c10000.mmc: mmc1 base at 0x0000000011c10000, max clock rate 133 MHz
[ 2.993112] mmc1: new ultra high speed SDR104 SDHC card at address aaaa
[ 3.014052] mmcblk1: mmc1:aaaa SC32G 29.7 GiB
[ 3.033826] mmcblk1: p1
[ 7.651447] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD52)
[ 12.768118] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD52)
[ 17.888240] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD0)
[ 23.008122] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD8)
[ 28.128098] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD5)
[ 33.248115] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD5)
[ 38.368119] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD5)
[ 43.488112] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD5)
[ 48.608110] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD55)
[ 53.728121] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD55)
[ 58.848115] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD55)
[ 63.968118] renesas_sdhi_internal_dmac 11c00000.mmc: timeout waiting for hardware interrupt (CMD55)
After reverting:
--------------
[ 2.463112] renesas_sdhi_internal_dmac 11c10000.mmc: mmc1 base at 0x0000000011c10000, max clock rate 133 MHz
[ 2.472569] rz-ssi-pcm-audio 10049c00.ssi: DMA enabled
[ 2.474124] renesas_sdhi_internal_dmac 11c00000.mmc: mmc0 base at 0x0000000011c00000, max clock rate 133 MHz
[ 2.600050] mmc0: new HS200 MMC card at address 0001
[ 2.608931] mmcblk0: mmc0:0001 G1M15M 59.3 GiB
[ 2.623432] mmcblk0: p1
[ 2.629863] mmcblk0boot0: mmc0:0001 G1M15M 31.5 MiB
[ 2.640008] mmcblk0boot1: mmc0:0001 G1M15M 31.5 MiB
[ 2.650212] mmcblk0rpmb: mmc0:0001 G1M15M 4.00 MiB, chardev (242:0)
[ 2.681609] Microchip KSZ9131 Gigabit PHY 11c20000.ethernet-ffffffff:07: attached PHY driver (mii_bus:phy_addr=11c20000.ethernet-ffffffff:07, irq=54)
[ 3.020085] mmc1: new ultra high speed SDR104 SDHC card at address aaaa
[ 3.033079] mmcblk1: mmc1:aaaa SC32G 29.7 GiB
[ 3.048402] mmcblk1: p1
Cheers,
Biju
> -----Original Message-----
> From: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>
> Sent: Tuesday, June 27, 2023 6:20 PM
> To: Dennis Zhou <dennis@xxxxxxxxxx>
> Cc: Ulf Hansson <ulf.hansson@xxxxxxxxxx>; Linux MMC List <linux-
> mmc@xxxxxxxxxxxxxxx>; Linux Kernel Mailing List <linux-
> kernel@xxxxxxxxxxxxxxx>; Biju Das <biju.das.jz@xxxxxxxxxxxxxx>; Wolfram
> Sang <wsa+renesas@xxxxxxxxxxxxxxxxxxxx>; Linux-Renesas <linux-renesas-
> soc@xxxxxxxxxxxxxxx>
> Subject: Re: [PATCH v2] mmc: inline the first mmc_scan() on
> mmc_start_host()
>
> Hi Dennis,
>
> On Thu, Mar 30, 2023 at 1:48 AM Dennis Zhou <dennis@xxxxxxxxxx> wrote:
> > When using dm-verity with a data partition on an emmc device,
> > dm-verity races with the discovery of attached emmc devices. This is
> > because mmc's probing code sets up the host data structure then a work
> > item is scheduled to do discovery afterwards. To prevent this race on
> > init, let's inline the first call to detection, __mm_scan(), and let
> > subsequent detect calls be handled via the workqueue.
> >
> > Signed-off-by: Dennis Zhou <dennis@xxxxxxxxxx>
>
> Thanks for your patch, which is now commit 2cc83bf7d41113d9 ("mmc:
> core: Allow mmc_start_host() synchronously detect a card") in linux-
> next/master mmc/next next-20230614 next-20230615 next-20230616
>
> I have bisected the following failure on Renesas Salvator-XS with R-Car
> H3
> ES2.0 to the above commit:
>
> renesas_sdhi_internal_dmac ee140000.mmc: timeout waiting for
> hardware interrupt (CMD0)
> renesas_sdhi_internal_dmac ee140000.mmc: timeout waiting for
> hardware interrupt (CMD1)
> renesas_sdhi_internal_dmac ee140000.mmc: timeout waiting for
> hardware interrupt (CMD0)
> renesas_sdhi_internal_dmac ee140000.mmc: timeout waiting for
> hardware interrupt (CMD1)
> mmc0: Failed to initialize a non-removable card
>
> Reverting the commit fixes the issue for me.
>
> > --- a/drivers/mmc/core/core.c
> > +++ b/drivers/mmc/core/core.c
> > @@ -2185,10 +2185,8 @@ int mmc_card_alternative_gpt_sector(struct
> > mmc_card *card, sector_t *gpt_sector) }
> > EXPORT_SYMBOL(mmc_card_alternative_gpt_sector);
> >
> > -void mmc_rescan(struct work_struct *work)
> > +static void __mmc_rescan(struct mmc_host *host)
> > {
> > - struct mmc_host *host =
> > - container_of(work, struct mmc_host, detect.work);
> > int i;
> >
> > if (host->rescan_disable)
> > @@ -2249,6 +2247,14 @@ void mmc_rescan(struct work_struct *work)
> > mmc_schedule_delayed_work(&host->detect, HZ); }
> >
> > +void mmc_rescan(struct work_struct *work) {
> > + struct mmc_host *host =
> > + container_of(work, struct mmc_host, detect.work);
> > +
> > + __mmc_rescan(host);
> > +}
> > +
> > void mmc_start_host(struct mmc_host *host) {
> > host->f_init = max(min(freqs[0], host->f_max), host->f_min);
> > @@ -2261,7 +2267,8 @@ void mmc_start_host(struct mmc_host *host)
> > }
> >
> > mmc_gpiod_request_cd_irq(host);
> > - _mmc_detect_change(host, 0, false);
> > + host->detect_change = 1;
> > + __mmc_rescan(host);
> > }
> >
> > void __mmc_stop_host(struct mmc_host *host)
>
> Gr{oetje,eeting}s,
>
> Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-
> m68k.org
>
> In personal conversations with technical people, I call myself a hacker.
> But when I'm talking to journalists I just say "programmer" or something
> like that.
> -- Linus Torvalds