mirror of
https://github.com/openwrt/openwrt.git
synced 2024-12-23 07:22:33 +00:00
ca5c695a45
Deleted following upstreamed patches: bcm27xx: 950-0006-drm-vc4-hdmi-Fix-HPD-GPIO-detection.patch bcm27xx: 950-0420-drm-vc4-Adopt-the-dma-configuration-from-the-HVS-or-.patch bcm27xx: 950-0425-drm-vc4-A-present-but-empty-dmas-disables-audio.patch bcm27xx: 950-0432-vc4-drm-Avoid-full-hdmi-audio-fifo-writes.patch bcm27xx: 950-0433-vc4-drm-vc4_plane-Remove-subpixel-positioning-check.patch bcm27xx: 950-0435-drm-vc4-Correct-pixel-order-for-DSI0.patch bcm27xx: 950-0436-drm-vc4-Register-dsi0-as-the-correct-vc4-encoder-typ.patch bcm27xx: 950-0437-drm-vc4-Fix-dsi0-interrupt-support.patch bcm27xx: 950-0438-drm-vc4-Add-correct-stop-condition-to-vc4_dsi_encode.patch bcm27xx: 950-0443-drm-vc4-Fix-timings-for-interlaced-modes.patch bcm27xx: 950-0445-drm-vc4-Fix-margin-calculations-for-the-right-bottom.patch bcm27xx: 950-0475-drm-vc4-Reset-HDMI-MISC_CONTROL-register.patch bcm27xx: 950-0476-drm-vc4-Release-workaround-buffer-and-DMA-in-error-p.patch bcm27xx: 950-0477-drm-vc4-Correct-DSI-divider-calculations.patch bcm27xx: 950-0664-drm-vc4-dsi-Correct-max-divider-to-255-not-7.patch bcm53xx: 072-next-ARM_dts_BCM53015-add-mr26.patch mediatek: 920-linux-next-dts-mt7622-bpi-r64-fix-wps-button.patch Manually rebased following patches: bcm27xx: 950-0004-drm-vc4-hdmi-Remove-the-DDC-probing-for-status-detec.patch bcm27xx: 950-0700-net-phy-lan87xx-Decrease-phy-polling-rate.patch bcm27xx: 950-0711-drm-vc4-Rename-bridge-to-out_bridge.patch bcm27xx: 950-0713-drm-vc4-Remove-splitting-the-bridge-chain-from-the-d.patch bcm27xx: 950-0715-drm-vc4-Convert-vc4_dsi-to-using-a-bridge-instead-of.patch bcm27xx: 950-0787-vc4-drm-vc4_plane-Keep-fractional-source-coords-insi.patch bcm27xx: 950-0914-mmc-block-Don-t-do-single-sector-reads-during-recove.patch Runtime tested on turris-omnia and glinet-b1300. Tested-by: John Audia <therealgraysky@proton.me> [bcm2711/RPi4B, mt7622/RT3200] Signed-off-by: Petr Štetiar <ynezz@true.cz>
39 lines
1.4 KiB
Diff
39 lines
1.4 KiB
Diff
From b70da899940777d92338d8f83f7a6236818ad08a Mon Sep 17 00:00:00 2001
|
|
From: Jonathan Bell <jonathan@raspberrypi.com>
|
|
Date: Mon, 16 May 2022 10:28:27 +0100
|
|
Subject: [PATCH] mmc: block: Don't do single-sector reads during
|
|
recovery
|
|
|
|
See https://github.com/raspberrypi/linux/issues/5019
|
|
|
|
If an SD card has degraded performance such that IO operations time out
|
|
then the MMC block layer will leak SG DMA mappings in the swiotlb during
|
|
recovery. It retries the same SG and this causes the leak, as it is
|
|
mapped twice - once in sdhci_pre_req() and again during single-block
|
|
reads in sdhci_prepare_data().
|
|
|
|
Resetting the card (including power-cycling if a regulator for vmmc is
|
|
present) ought to be enough to recover a stuck state, so for now don't
|
|
try single-block reads in the recovery path.
|
|
|
|
Signed-off-by: Jonathan Bell <jonathan@raspberrypi.com>
|
|
---
|
|
drivers/mmc/core/block.c | 6 +++++-
|
|
1 file changed, 5 insertions(+), 1 deletion(-)
|
|
|
|
--- a/drivers/mmc/core/block.c
|
|
+++ b/drivers/mmc/core/block.c
|
|
@@ -1870,7 +1870,11 @@ static void mmc_blk_mq_rw_recovery(struc
|
|
return;
|
|
}
|
|
|
|
- if (rq_data_dir(req) == READ && brq->data.blocks >
|
|
+ /*
|
|
+ * XXX: don't do single-sector reads, as it leaks a SG DMA
|
|
+ * mapping when reusing the still-pending req.
|
|
+ */
|
|
+ if (0 && rq_data_dir(req) == READ && brq->data.blocks >
|
|
queue_physical_block_size(mq->queue) >> 9) {
|
|
/* Read one (native) sector at a time */
|
|
mmc_blk_read_single(mq, req);
|