From mboxrd@z Thu Jan 1 00:00:00 1970 From: Seung-Woo Kim Subject: Re: [PATCH] mmc: dw_mmc: remove UBSAN warning in dw_mci_setup_bus() Date: Fri, 10 Jun 2016 10:29:25 +0900 Message-ID: <575A17F5.1050609@samsung.com> References: <1465358840-22826-1-git-send-email-sw0312.kim@samsung.com> <5759633E.7090706@samsung.com> Reply-To: sw0312.kim@samsung.com Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-reply-to: <5759633E.7090706@samsung.com> Sender: linux-kernel-owner@vger.kernel.org To: Jaehoon Chung Cc: ulf.hansson@linaro.org, linux-mmc@vger.kernel.org, linux-kernel@vger.kernel.org, Seung-Woo Kim List-Id: linux-mmc@vger.kernel.org Hi Jaehoon, On 2016=EB=85=84 06=EC=9B=94 09=EC=9D=BC 21:38, Jaehoon Chung wrote: > Hi Seung-Woo, >=20 > On 06/08/2016 01:07 PM, Seung-Woo Kim wrote: >> This patch removes following UBSAN warnings in dw_mci_setup_bus(). >> The warnings are caused because of shift with more than 31 on 32 >> bit variable, so this patch fixes to shift only for less than 32. >> >> UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1102:14 >> shift exponent 250 is too large for 32-bit type 'unsigned int' >> Call trace: >> [] dump_backtrace+0x0/0x380 >> [] show_stack+0x14/0x20 >> [] dump_stack+0xe0/0x120 >> [] ubsan_epilogue+0x18/0x68 >> [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1b= c >> [] dw_mci_setup_bus+0x3a0/0x438 >> [...] >> >> UBSAN: Undefined behaviour in drivers/mmc/host/dw_mmc.c:1132:27 >> shift exponent 250 is too large for 32-bit type 'unsigned int' >> Call trace: >> [] dump_backtrace+0x0/0x380 >> [] show_stack+0x14/0x20 >> [] dump_stack+0xe0/0x120 >> [] ubsan_epilogue+0x18/0x68 >> [] __ubsan_handle_shift_out_of_bounds+0x18c/0x1b= c >> [] dw_mci_setup_bus+0x384/0x438 >=20 > The below message can be discarded. You are right, like above message, below part can be replaced as [...]. >=20 >> [] dw_mci_set_ios+0x184/0x798 >> [] mmc_power_up+0x11c/0x260 >> [] mmc_start_host+0x88/0x100 >> [] mmc_add_host+0x6c/0x128 >> [] dw_mci_probe+0x1088/0x1750 >> [] dw_mci_pltfm_register+0x108/0x178 >> [] dw_mci_exynos_probe+0x4c/0x88 >> [] platform_drv_probe+0x78/0x180 >> [] driver_probe_device+0x144/0x460 >> [] __driver_attach+0xf4/0x140 >> [] bus_for_each_dev+0xf0/0x160 >> [] driver_attach+0x34/0x58 >> [] bus_add_driver+0x2c0/0x398 >> [] driver_register+0xbc/0x1e0 >> [] __platform_driver_register+0x84/0xa8 >> [] dw_mci_exynos_pltfm_driver_init+0x18/0x20 >> [] do_one_initcall+0xa0/0x2c8 >> [] kernel_init_freeable+0x52c/0x5dc >> [] kernel_init+0x1c/0xf8 >> [] ret_from_fork+0x10/0x40 >> >> Signed-off-by: Seung-Woo Kim >> --- >> drivers/mmc/host/dw_mmc.c | 5 +++-- >> 1 files changed, 3 insertions(+), 2 deletions(-) >> >> diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c >> index 2cc6123..dff045e 100644 >> --- a/drivers/mmc/host/dw_mmc.c >> +++ b/drivers/mmc/host/dw_mmc.c >> @@ -1099,7 +1099,8 @@ static void dw_mci_setup_bus(struct dw_mci_slo= t *slot, bool force_clkinit) >> =20 >> div =3D (host->bus_hz !=3D clock) ? DIV_ROUND_UP(div, 2) : 0; >> =20 >> - if ((clock << div) !=3D slot->__clk_old || force_clkinit) >> + if (((div < 32) ? (clock << div) : 0) !=3D slot->__clk_old || >=20 > Well, we don't expect that clock is 0. > if clock is 0, it should be passed to " if (!clock)".. >=20 > During initializing card, clock is 400KHz or less..I understood what = you want to fix. > But I taught this is not correct. It seems slot->__clk_old is not really clock but a kind of value to store both clock and div in one variable. And at least, my compiler calculates right shift with more than 32 as 0. I am not sure there is other proper value for the case. By the way, in my test environment, clock is calculated as like following steps: mmc_host mmc0: Bus speed (slot 0) =3D 200000000Hz (slot req 400000Hz, actual 400000HZ div =3D 250) mmc_host mmc0: Bus speed (slot 0) =3D 200000000Hz (slot req 200000000Hz= , actual 200000000HZ div =3D 0) mmc_host mmc0: Bus speed (slot 0) =3D 200000000Hz (slot req 52000000Hz, actual 50000000HZ div =3D 2) mmc_host mmc0: Bus speed (slot 0) =3D 400000000Hz (slot req 52000000Hz, actual 50000000HZ div =3D 4) mmc_host mmc0: Bus speed (slot 0) =3D 400000000Hz (slot req 200000000Hz= , actual 200000000HZ div =3D 1) and only the first case is reported as the warning. If you let me know any proper value, then I will fix with the value. Thanks, - Seung-Woo Kim >=20 >=20 >> + force_clkinit) >> dev_info(&slot->mmc->class_dev, >> "Bus speed (slot %d) =3D %dHz (slot req %dHz, actual %dHZ div = =3D %d)\n", >> slot->id, host->bus_hz, clock, >> @@ -1129,7 +1130,7 @@ static void dw_mci_setup_bus(struct dw_mci_slo= t *slot, bool force_clkinit) >> mci_send_cmd(slot, sdmmc_cmd_bits, 0); >> =20 >> /* keep the clock with reflecting clock dividor */ >> - slot->__clk_old =3D clock << div; >> + slot->__clk_old =3D (div < 32) ? (clock << div) : 0; >=20 > Also, clock is not 0.. >=20 >=20 > Best Regards, > Jaehoon Chung >=20 >> } >> =20 >> host->current_speed =3D clock; >> >=20 >=20 >=20 --=20 Seung-Woo Kim Samsung Software R&D Center