linux-kernel - [PATCH 0/6] Preparations for ASIC3 support in the TMIO MMC driver

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-next>] [day] [month] [year] [list]

Date:	Mon, 25 May 2009 21:27:56 +0200
From:	Philipp Zabel <philipp.zabel@...il.com>
To:	linux-kernel@...r.kernel.org
Cc:	Ian Molton <spyro@....com>
Subject: [PATCH 0/6] Preparations for ASIC3 support in the TMIO MMC driver

Hi,

this series is really only two parts preparation for ASIC3:
 [PATCH 1/5] MFD,mmc: tmio_mmc: make HCLK configurable
 [PATCH 2/5] mmc: tmio_mmc: add bus_shift support
and three parts cleanup:
 [PATCH 3/5] mmc: tmio_mmc: don't use set_irq_type
 [PATCH 4/5] mmc: tmio_mmc: correct probe return value for num_resources != 3
 [PATCH 5/5] mmc: tmio_mmc: move probe function into __init section

I have still a few remaining differences until tmio_mmc can support ASIC3,
but as patch 2/5 causes changes all over the place, I'd like to get feedback
on that before piling patches on top of it.

I also have a question about the clock divider settings. Currently,
tmio_mmc_set_clock returns the wrong divider setting for new_clock=f_min
(0x40 instead of 0x80). I tried to fix it like below:

+#define TMIO_CARD_CLOCK_ENABLE 0x0100
+#define TMIO_CARD_CLOCK_FOR_SD 0x8000
+
-static void tmio_mmc_set_clock(struct tmio_mmc_host *host, int new_clock)
+static void tmio_mmc_set_clock(struct tmio_mmc_host *host, int new_clock)
 {
        u32 clk = 0, clock;
 
        if (new_clock) {
                for (clock = host->mmc->f_min, clk = 0x100; new_clock >= (clock<<1); ) {
                        clock <<= 1;
                        clk >>= 1;
                }
-               if (clk & 0x1)
-                       clk = 0x20000;
+               /* For fastest speed we disable the divider. */
+               sd_config_write8(host, CNF_SD_CLK_MODE, (clk == 0) ? 0 : 1);
+               clk >>= 1;
 
-               clk >>= 2;
-               sd_config_write8(host, CNF_SD_CLK_MODE, (clk & 0x8000) ? 0 : 1);
-               clk |= 0x100;
+               clk |= host->clock_for_sd | TMIO_CARD_CLOCK_ENABLE;
        }
 
        sd_ctrl_write16(host, CTL_SD_CARD_CLK_CTL, clk);

But the ASIC3 code was handling the 0x8000 bit (called FOR_SD in the ASIC3 headers)
of CTL_SD_CARD_CLK_CTL differently - there it was enabled/disabled based on 1-bit
or 4-bit MMC mode. I have no idea what it is good for - is there any TMIO documentation
available that includes this information?

regards
Philipp
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/