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 PHC | |
Open Source and information security mailing list archives
| ||
|
Date: Tue, 29 Dec 2020 13:27:40 +0800 From: Xu Yilun <yilun.xu@...el.com> To: broonie@...nel.org, linux-spi@...r.kernel.org Cc: trix@...hat.com, lgoncalv@...hat.com, yilun.xu@...el.com, hao.wu@...el.com, matthew.gerlach@...ux.intel.com, russell.h.weight@...el.com, linux-kernel@...r.kernel.org Subject: [PATCH 0/2] fix the issue when xfer by spi-altera When doing spi xfer by spi-altera, divide by 0 exception happens in spi_transfer_wait(), This is because the xfer->speed_hz is always clamped to 0 by spi->controller->max_speed_hz, the feature is introduced in: commit 9326e4f1e5dd ("spi: Limit the spi device max speed to controller's max speed") The spi-altera doesn't have hardware indication for controller's max_speed_hz, so its value is uninitialized as 0. Patch #1 fixes the issue of spi_altera driver. When doing polling mode xfer, its transfer_one() callback should return 1, to indicate the xfer is finished. It should return 0 for irq mode xfer. With this patch the polling mode xfer is OK as it needs no spi_transfer_wait() anymore. But the irq mode xfer is still broken. So Patch #2 assumes 1khz xfer speed if the xfer->speed_hz is not assigned. I try to avoid the divide by 0 issue and ensures a reasonable tolerant waiting time in a generic way. Xu Yilun (2): spi: altera: fix return value for altera_spi_txrx() spi: fix the divide by 0 error when calculating xfer waiting time drivers/spi/spi-altera.c | 26 ++++++++++++++------------ drivers/spi/spi.c | 4 +++- 2 files changed, 17 insertions(+), 13 deletions(-) -- 2.7.4
Powered by blists - more mailing lists