summaryrefslogtreecommitdiff
path: root/tools/testing/selftests/kvm/lib/kvm_util.c
diff options
context:
space:
mode:
authorWolfram Sang <wsa+renesas@sang-engineering.com>2018-06-01 14:00:37 +0300
committerUlf Hansson <ulf.hansson@linaro.org>2018-06-04 12:16:58 +0300
commitef5332c10d4f332a2ac79e9ad5452f4e89d1815a (patch)
tree454be2b7b20be91cdf6a330da99a8e5cd7432336 /tools/testing/selftests/kvm/lib/kvm_util.c
parent77252da7cd2cd6d6eed31aaa8d69b05e271bf455 (diff)
downloadlinux-ef5332c10d4f332a2ac79e9ad5452f4e89d1815a.tar.xz
mmc: renesas_sdhi: really fix WP logic regressions
This reverts commit e060d376cc61 ("mmc: renesas_sdhi: fix WP detection") and adds some code to really fix the regressions. It was missed so far that Renesas R-Car instantiations of SDHI chose to disable internal WP and used the existence of "wp-gpios" to en/disable WP at all. With the first refactoring by Yamada-san with commit 2ad1db059b9a ("mmc: renesas_sdhi: use MMC_CAP2_NO_WRITE_PROTECT instead of TMIO own flag"), WP was always disabled even when GPIOs were present. With Wolfram's first fix which gets now reverted, GPIOs were honored. But when not available, the fallback was to internal WP and not to disabled WP. This caused wrong WP status on uSD card slots. Restore the old behaviour now. By default, WP is disabled. When a GPIO is found, the GPIO re-enables WP. We will think about possible better ways to handle this in the future. Tested on a previously regressing Renesas Lager board (H2) and a still working Renesas Salvator-X board (M3-W). Reported-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com> Cc: stable@vger.kernel.org # v4.17+ Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Diffstat (limited to 'tools/testing/selftests/kvm/lib/kvm_util.c')
0 files changed, 0 insertions, 0 deletions