Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1328154 > unrolled thread
| Started by | Alexey Khoroshilov <khoroshilov@ispras.ru> |
|---|---|
| First post | 2016-02-06 00:40 +0100 |
| Last post | 2016-02-08 16:30 +0100 |
| Articles | 2 — 2 participants |
Back to article view | Back to linux.kernel
[PATCH] mmc: mmc_spi: add checks for dma mapping error Alexey Khoroshilov <khoroshilov@ispras.ru> - 2016-02-06 00:40 +0100
Re: [PATCH] mmc: mmc_spi: add checks for dma mapping error Ulf Hansson <ulf.hansson@linaro.org> - 2016-02-08 16:30 +0100
| From | Alexey Khoroshilov <khoroshilov@ispras.ru> |
|---|---|
| Date | 2016-02-06 00:40 +0100 |
| Subject | [PATCH] mmc: mmc_spi: add checks for dma mapping error |
| Message-ID | <qYY8P-75F-29@gated-at.bofh.it> |
There is no checks for dma mapping errors in mmc_spi.
Tha patch fixes that and by the way it adds dma_unmap_single(ones_dma)
that was left on a failure path mmc_spi_probe().
Found by Linux Driver Verification project (linuxtesting.org).
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
---
drivers/mmc/host/mmc_spi.c | 15 +++++++++++++--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 1c1b45ef3faf..3446097a43c0 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -925,6 +925,10 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd,
dma_addr = dma_map_page(dma_dev, sg_page(sg), 0,
PAGE_SIZE, dir);
+ if (dma_mapping_error(dma_dev, dma_addr)) {
+ data->error = -EFAULT;
+ break;
+ }
if (direction == DMA_TO_DEVICE)
t->tx_dma = dma_addr + sg->offset;
else
@@ -1393,10 +1397,12 @@ static int mmc_spi_probe(struct spi_device *spi)
host->dma_dev = dev;
host->ones_dma = dma_map_single(dev, ones,
MMC_SPI_BLOCKSIZE, DMA_TO_DEVICE);
+ if (dma_mapping_error(dev, host->ones_dma))
+ goto fail_ones_dma;
host->data_dma = dma_map_single(dev, host->data,
sizeof(*host->data), DMA_BIDIRECTIONAL);
-
- /* REVISIT in theory those map operations can fail... */
+ if (dma_mapping_error(dev, host->data_dma))
+ goto fail_data_dma;
dma_sync_single_for_cpu(host->dma_dev,
host->data_dma, sizeof(*host->data),
@@ -1462,6 +1468,11 @@ fail_glue_init:
if (host->dma_dev)
dma_unmap_single(host->dma_dev, host->data_dma,
sizeof(*host->data), DMA_BIDIRECTIONAL);
+fail_data_dma:
+ if (host->dma_dev)
+ dma_unmap_single(host->dma_dev, host->ones_dma,
+ MMC_SPI_BLOCKSIZE, DMA_TO_DEVICE);
+fail_ones_dma:
kfree(host->data);
fail_nobuf1:
--
1.9.1
[toc] | [next] | [standalone]
| From | Ulf Hansson <ulf.hansson@linaro.org> |
|---|---|
| Date | 2016-02-08 16:30 +0100 |
| Message-ID | <qZVVh-782-39@gated-at.bofh.it> |
| In reply to | #1328154 |
On 6 February 2016 at 00:36, Alexey Khoroshilov <khoroshilov@ispras.ru> wrote:
> There is no checks for dma mapping errors in mmc_spi.
> Tha patch fixes that and by the way it adds dma_unmap_single(ones_dma)
> that was left on a failure path mmc_spi_probe().
>
> Found by Linux Driver Verification project (linuxtesting.org).
>
> Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Thanks, applied for fixes!
Kind regards
Uffe
> ---
> drivers/mmc/host/mmc_spi.c | 15 +++++++++++++--
> 1 file changed, 13 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
> index 1c1b45ef3faf..3446097a43c0 100644
> --- a/drivers/mmc/host/mmc_spi.c
> +++ b/drivers/mmc/host/mmc_spi.c
> @@ -925,6 +925,10 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd,
>
> dma_addr = dma_map_page(dma_dev, sg_page(sg), 0,
> PAGE_SIZE, dir);
> + if (dma_mapping_error(dma_dev, dma_addr)) {
> + data->error = -EFAULT;
> + break;
> + }
> if (direction == DMA_TO_DEVICE)
> t->tx_dma = dma_addr + sg->offset;
> else
> @@ -1393,10 +1397,12 @@ static int mmc_spi_probe(struct spi_device *spi)
> host->dma_dev = dev;
> host->ones_dma = dma_map_single(dev, ones,
> MMC_SPI_BLOCKSIZE, DMA_TO_DEVICE);
> + if (dma_mapping_error(dev, host->ones_dma))
> + goto fail_ones_dma;
> host->data_dma = dma_map_single(dev, host->data,
> sizeof(*host->data), DMA_BIDIRECTIONAL);
> -
> - /* REVISIT in theory those map operations can fail... */
> + if (dma_mapping_error(dev, host->data_dma))
> + goto fail_data_dma;
>
> dma_sync_single_for_cpu(host->dma_dev,
> host->data_dma, sizeof(*host->data),
> @@ -1462,6 +1468,11 @@ fail_glue_init:
> if (host->dma_dev)
> dma_unmap_single(host->dma_dev, host->data_dma,
> sizeof(*host->data), DMA_BIDIRECTIONAL);
> +fail_data_dma:
> + if (host->dma_dev)
> + dma_unmap_single(host->dma_dev, host->ones_dma,
> + MMC_SPI_BLOCKSIZE, DMA_TO_DEVICE);
> +fail_ones_dma:
> kfree(host->data);
>
> fail_nobuf1:
> --
> 1.9.1
>
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web