[PATCH 0/4] Fix the parameter of dma_map_sg()
From: Xiang Chen <chenxiang66@hisilicon.com> According to Documentation/core-api/dma-api-howto.rst, the parameters of dma_unmap_sg() must be the same as those which are passed in to the scatter/gather mapping API. But for some drivers under crypto, the <nents> parameter of dma_unmap_sg() is number of elements after mapping. So fix them. Part of the document is as follows: To unmap a scatterlist, just call:: dma_unmap_sg(dev, sglist, nents, direction); Again, make sure DMA activity has already finished. .. note:: The 'nents' argument to the dma_unmap_sg call must be the _same_ one you passed into the dma_map_sg call, it should _NOT_ be the 'count' value _returned_ from the dma_map_sg call. chenxiang (4): crypto: amlogic - Fix the parameter of dma_unmap_sg() crypto: cavium - Fix the parameter of dma_unmap_sg() crypto: ux500 - Fix the parameter of dma_unmap_sg() crypto: allwinner - Fix the parameter of dma_unmap_sg() drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c | 3 ++- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c | 3 ++- drivers/crypto/amlogic/amlogic-gxl-cipher.c | 6 +++--- drivers/crypto/cavium/nitrox/nitrox_reqmgr.c | 8 ++++---- drivers/crypto/ux500/cryp/cryp_core.c | 4 ++-- drivers/crypto/ux500/hash/hash_core.c | 2 +- 8 files changed, 26 insertions(+), 18 deletions(-) -- 2.8.1
From: Xiang Chen <chenxiang66@hisilicon.com> For function dma_unmap_sg(), the <nents> parameter should be number of elements in the scatterlist prior to the mapping, not after the mapping. So fix this usage. Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> --- drivers/crypto/amlogic/amlogic-gxl-cipher.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/crypto/amlogic/amlogic-gxl-cipher.c b/drivers/crypto/amlogic/amlogic-gxl-cipher.c index 8b5e073..c6865cbd 100644 --- a/drivers/crypto/amlogic/amlogic-gxl-cipher.c +++ b/drivers/crypto/amlogic/amlogic-gxl-cipher.c @@ -236,10 +236,10 @@ static int meson_cipher(struct skcipher_request *areq) dma_unmap_single(mc->dev, phykeyiv, keyivlen, DMA_TO_DEVICE); if (areq->src == areq->dst) { - dma_unmap_sg(mc->dev, areq->src, nr_sgs, DMA_BIDIRECTIONAL); + dma_unmap_sg(mc->dev, areq->src, sg_nents(areq->src), DMA_BIDIRECTIONAL); } else { - dma_unmap_sg(mc->dev, areq->src, nr_sgs, DMA_TO_DEVICE); - dma_unmap_sg(mc->dev, areq->dst, nr_sgd, DMA_FROM_DEVICE); + dma_unmap_sg(mc->dev, areq->src, sg_nents(areq->src), DMA_TO_DEVICE); + dma_unmap_sg(mc->dev, areq->dst, sg_nents(areq->dst), DMA_FROM_DEVICE); } if (areq->iv && ivsize > 0) { -- 2.8.1
From: Xiang Chen <chenxiang66@hisilicon.com> For function dma_unmap_sg(), the <nents> parameter should be number of elements in the scatterlist prior to the mapping, not after the mapping. So fix this usage. Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> --- drivers/crypto/cavium/nitrox/nitrox_reqmgr.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c index 53ef067..1263194 100644 --- a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c +++ b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c @@ -170,7 +170,7 @@ static int dma_map_inbufs(struct nitrox_softreq *sr, sr->in.total_bytes += sg_dma_len(sg); sr->in.sg = req->src; - sr->in.sgmap_cnt = nents; + sr->in.sgmap_cnt = sg_nents(req->src); ret = create_sg_component(sr, &sr->in, sr->in.sgmap_cnt); if (ret) goto incomp_err; @@ -178,7 +178,7 @@ static int dma_map_inbufs(struct nitrox_softreq *sr, return 0; incomp_err: - dma_unmap_sg(dev, req->src, nents, DMA_BIDIRECTIONAL); + dma_unmap_sg(dev, req->src, sg_nents(req->src), DMA_BIDIRECTIONAL); sr->in.sgmap_cnt = 0; return ret; } @@ -195,7 +195,7 @@ static int dma_map_outbufs(struct nitrox_softreq *sr, return -EINVAL; sr->out.sg = req->dst; - sr->out.sgmap_cnt = nents; + sr->out.sgmap_cnt = sg_nents(req->dst); ret = create_sg_component(sr, &sr->out, sr->out.sgmap_cnt); if (ret) goto outcomp_map_err; @@ -203,7 +203,7 @@ static int dma_map_outbufs(struct nitrox_softreq *sr, return 0; outcomp_map_err: - dma_unmap_sg(dev, req->dst, nents, DMA_BIDIRECTIONAL); + dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_BIDIRECTIONAL); sr->out.sgmap_cnt = 0; sr->out.sg = NULL; return ret; -- 2.8.1
On Tue, Feb 09, 2021 at 02:59:23PM +0800, chenxiang wrote:
From: Xiang Chen <chenxiang66@hisilicon.com>
For function dma_unmap_sg(), the <nents> parameter should be number of elements in the scatterlist prior to the mapping, not after the mapping. So fix this usage.
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> --- drivers/crypto/cavium/nitrox/nitrox_reqmgr.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c index 53ef067..1263194 100644 --- a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c +++ b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c @@ -170,7 +170,7 @@ static int dma_map_inbufs(struct nitrox_softreq *sr, sr->in.total_bytes += sg_dma_len(sg);
sr->in.sg = req->src; - sr->in.sgmap_cnt = nents; + sr->in.sgmap_cnt = sg_nents(req->src); ret = create_sg_component(sr, &sr->in, sr->in.sgmap_cnt);
So you're changing the count passed to create_sg_component. Are you sure that's correct? Even if it is correct you should change your patch description to document this change. Thanks, -- Email: Herbert Xu <herbert@gondor.apana.org.au> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Hi Herbert, 在 2021/3/3 16:59, Herbert Xu 写道:
On Tue, Feb 09, 2021 at 02:59:23PM +0800, chenxiang wrote:
From: Xiang Chen <chenxiang66@hisilicon.com>
For function dma_unmap_sg(), the <nents> parameter should be number of elements in the scatterlist prior to the mapping, not after the mapping. So fix this usage.
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> --- drivers/crypto/cavium/nitrox/nitrox_reqmgr.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c index 53ef067..1263194 100644 --- a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c +++ b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c @@ -170,7 +170,7 @@ static int dma_map_inbufs(struct nitrox_softreq *sr, sr->in.total_bytes += sg_dma_len(sg);
sr->in.sg = req->src; - sr->in.sgmap_cnt = nents; + sr->in.sgmap_cnt = sg_nents(req->src); ret = create_sg_component(sr, &sr->in, sr->in.sgmap_cnt); So you're changing the count passed to create_sg_component. Are you sure that's correct? Even if it is correct you should change your patch description to document this change.
Thank you for reminding me about this. I didn't notice that it changes the count passed to create_sg_component. I have a change on this patch as follows, and please have a check on it: --- a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c +++ b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c @@ -58,14 +58,14 @@ static void softreq_unmap_sgbufs(struct nitrox_softreq *sr) struct device *dev = DEV(ndev); - dma_unmap_sg(dev, sr->in.sg, sr->in.sgmap_cnt, DMA_BIDIRECTIONAL); + dma_unmap_sg(dev, sr->in.sg, sg_nents(sr->in.sg), DMA_BIDIRECTIONAL); dma_unmap_single(dev, sr->in.sgcomp_dma, sr->in.sgcomp_len, DMA_TO_DEVICE); kfree(sr->in.sgcomp); sr->in.sg = NULL; sr->in.sgmap_cnt = 0; - dma_unmap_sg(dev, sr->out.sg, sr->out.sgmap_cnt, + dma_unmap_sg(dev, sr->out.sg, sg_nents(sr->out.sg), DMA_BIDIRECTIONAL); dma_unmap_single(dev, sr->out.sgcomp_dma, sr->out.sgcomp_len, DMA_TO_DEVICE); @@ -178,7 +178,8 @@ static int dma_map_inbufs(struct nitrox_softreq *sr, return 0; incomp_err: - dma_unmap_sg(dev, req->src, nents, DMA_BIDIRECTIONAL); + dma_unmap_sg(dev, req->src, sg_nents(req->src), + DMA_BIDIRECTIONAL); sr->in.sgmap_cnt = 0; return ret; } @@ -203,7 +204,8 @@ static int dma_map_outbufs(struct nitrox_softreq *sr, return 0; outcomp_map_err: - dma_unmap_sg(dev, req->dst, nents, DMA_BIDIRECTIONAL); + dma_unmap_sg(dev, req->dst, sg_nents(req->dst), + DMA_BIDIRECTIONAL); sr->out.sgmap_cnt = 0; sr->out.sg = NULL; return ret;
On Thu, Mar 04, 2021 at 10:18:51AM +0800, chenxiang (M) wrote:
Thank you for reminding me about this. I didn't notice that it changes the count passed to create_sg_component. I have a change on this patch as follows, and please have a check on it:
Yes this looks fine. Thanks, -- Email: Herbert Xu <herbert@gondor.apana.org.au> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
From: Xiang Chen <chenxiang66@hisilicon.com> For function dma_unmap_sg(), the <nents> parameter should be number of elements in the scatterlist prior to the mapping, not after the mapping. So fix this usage. Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> --- drivers/crypto/ux500/cryp/cryp_core.c | 4 ++-- drivers/crypto/ux500/hash/hash_core.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/crypto/ux500/cryp/cryp_core.c b/drivers/crypto/ux500/cryp/cryp_core.c index c3adeb2..9abf00e 100644 --- a/drivers/crypto/ux500/cryp/cryp_core.c +++ b/drivers/crypto/ux500/cryp/cryp_core.c @@ -608,12 +608,12 @@ static void cryp_dma_done(struct cryp_ctx *ctx) chan = ctx->device->dma.chan_mem2cryp; dmaengine_terminate_all(chan); dma_unmap_sg(chan->device->dev, ctx->device->dma.sg_src, - ctx->device->dma.sg_src_len, DMA_TO_DEVICE); + ctx->device->dma.nents_src, DMA_TO_DEVICE); chan = ctx->device->dma.chan_cryp2mem; dmaengine_terminate_all(chan); dma_unmap_sg(chan->device->dev, ctx->device->dma.sg_dst, - ctx->device->dma.sg_dst_len, DMA_FROM_DEVICE); + ctx->device->dma.nents_dst, DMA_FROM_DEVICE); } static int cryp_dma_write(struct cryp_ctx *ctx, struct scatterlist *sg, diff --git a/drivers/crypto/ux500/hash/hash_core.c b/drivers/crypto/ux500/hash/hash_core.c index da284b0..67b1237 100644 --- a/drivers/crypto/ux500/hash/hash_core.c +++ b/drivers/crypto/ux500/hash/hash_core.c @@ -190,7 +190,7 @@ static void hash_dma_done(struct hash_ctx *ctx) chan = ctx->device->dma.chan_mem2hash; dmaengine_terminate_all(chan); dma_unmap_sg(chan->device->dev, ctx->device->dma.sg, - ctx->device->dma.sg_len, DMA_TO_DEVICE); + ctx->device->dma.nents, DMA_TO_DEVICE); } static int hash_dma_write(struct hash_ctx *ctx, -- 2.8.1
From: Xiang Chen <chenxiang66@hisilicon.com> For function dma_unmap_sg(), the <nents> parameter should be number of elements in the scatterlist prior to the mapping, not after the mapping. So fix this usage. Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> --- drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c | 3 ++- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c | 3 ++- 4 files changed, 16 insertions(+), 8 deletions(-) diff --git a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c index 33707a2..54ae8d1 100644 --- a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c +++ b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c @@ -240,11 +240,14 @@ static int sun8i_ce_cipher_prepare(struct crypto_engine *engine, void *async_req theend_sgs: if (areq->src == areq->dst) { - dma_unmap_sg(ce->dev, areq->src, nr_sgs, DMA_BIDIRECTIONAL); + dma_unmap_sg(ce->dev, areq->src, sg_nents(areq->src), + DMA_BIDIRECTIONAL); } else { if (nr_sgs > 0) - dma_unmap_sg(ce->dev, areq->src, nr_sgs, DMA_TO_DEVICE); - dma_unmap_sg(ce->dev, areq->dst, nr_sgd, DMA_FROM_DEVICE); + dma_unmap_sg(ce->dev, areq->src, sg_nents(areq->src), + DMA_TO_DEVICE); + dma_unmap_sg(ce->dev, areq->dst, sg_nents(areq->dst), + DMA_FROM_DEVICE); } theend_iv: diff --git a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c index 2f09a37..8819471 100644 --- a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c +++ b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c @@ -405,7 +405,8 @@ int sun8i_ce_hash_run(struct crypto_engine *engine, void *breq) err = sun8i_ce_run_task(ce, flow, crypto_tfm_alg_name(areq->base.tfm)); dma_unmap_single(ce->dev, addr_pad, j * 4, DMA_TO_DEVICE); - dma_unmap_sg(ce->dev, areq->src, nr_sgs, DMA_TO_DEVICE); + dma_unmap_sg(ce->dev, areq->src, sg_nents(areq->src), + DMA_TO_DEVICE); dma_unmap_single(ce->dev, addr_res, digestsize, DMA_FROM_DEVICE); diff --git a/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c b/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c index ed2a69f..f945750 100644 --- a/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c +++ b/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c @@ -232,10 +232,13 @@ static int sun8i_ss_cipher(struct skcipher_request *areq) theend_sgs: if (areq->src == areq->dst) { - dma_unmap_sg(ss->dev, areq->src, nr_sgs, DMA_BIDIRECTIONAL); + dma_unmap_sg(ss->dev, areq->src, sg_nents(areq->src), + DMA_BIDIRECTIONAL); } else { - dma_unmap_sg(ss->dev, areq->src, nr_sgs, DMA_TO_DEVICE); - dma_unmap_sg(ss->dev, areq->dst, nr_sgd, DMA_FROM_DEVICE); + dma_unmap_sg(ss->dev, areq->src, sg_nents(areq->src), + DMA_TO_DEVICE); + dma_unmap_sg(ss->dev, areq->dst, sg_nents(areq->dst), + DMA_FROM_DEVICE); } theend_iv: diff --git a/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c b/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c index 11cbcbc..55c06c3a 100644 --- a/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c +++ b/drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c @@ -432,7 +432,8 @@ int sun8i_ss_hash_run(struct crypto_engine *engine, void *breq) err = sun8i_ss_run_hash_task(ss, rctx, crypto_tfm_alg_name(areq->base.tfm)); dma_unmap_single(ss->dev, addr_pad, j * 4, DMA_TO_DEVICE); - dma_unmap_sg(ss->dev, areq->src, nr_sgs, DMA_TO_DEVICE); + dma_unmap_sg(ss->dev, areq->src, sg_nents(areq->src), + DMA_TO_DEVICE); dma_unmap_single(ss->dev, addr_res, digestsize, DMA_FROM_DEVICE); kfree(pad); -- 2.8.1
Le Tue, Feb 09, 2021 at 02:59:25PM +0800, chenxiang a écrit :
From: Xiang Chen <chenxiang66@hisilicon.com>
For function dma_unmap_sg(), the <nents> parameter should be number of elements in the scatterlist prior to the mapping, not after the mapping. So fix this usage.
Signed-off-by: Xiang Chen <chenxiang66@hisilicon.com> --- drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c | 3 ++- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c | 3 ++- 4 files changed, 16 insertions(+), 8 deletions(-)
Hello Acked-by: Corentin LABBE <clabbe.montjoie@gmail.com> Tested-by: Corentin LABBE <clabbe.montjoie@gmail.com> Tested-on: sun8i-a83t-bananapi-m3 Tested-on: sun50i-h6-pine-h64 Tested-on: sun50i-h6-orangepi-one-plus Thanks
Ping... 在 2021/2/9 14:59, chenxiang 写道:
From: Xiang Chen <chenxiang66@hisilicon.com>
According to Documentation/core-api/dma-api-howto.rst, the parameters of dma_unmap_sg() must be the same as those which are passed in to the scatter/gather mapping API. But for some drivers under crypto, the <nents> parameter of dma_unmap_sg() is number of elements after mapping. So fix them.
Part of the document is as follows:
To unmap a scatterlist, just call::
dma_unmap_sg(dev, sglist, nents, direction);
Again, make sure DMA activity has already finished.
.. note::
The 'nents' argument to the dma_unmap_sg call must be the _same_ one you passed into the dma_map_sg call, it should _NOT_ be the 'count' value _returned_ from the dma_map_sg call.
chenxiang (4): crypto: amlogic - Fix the parameter of dma_unmap_sg() crypto: cavium - Fix the parameter of dma_unmap_sg() crypto: ux500 - Fix the parameter of dma_unmap_sg() crypto: allwinner - Fix the parameter of dma_unmap_sg()
drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c | 3 ++- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-cipher.c | 9 ++++++--- drivers/crypto/allwinner/sun8i-ss/sun8i-ss-hash.c | 3 ++- drivers/crypto/amlogic/amlogic-gxl-cipher.c | 6 +++--- drivers/crypto/cavium/nitrox/nitrox_reqmgr.c | 8 ++++---- drivers/crypto/ux500/cryp/cryp_core.c | 4 ++-- drivers/crypto/ux500/hash/hash_core.c | 2 +- 8 files changed, 26 insertions(+), 18 deletions(-)
On Sat, Feb 20, 2021 at 09:51:17AM +0800, chenxiang (M) wrote:
Ping...
Please be patient. -- Email: Herbert Xu <herbert@gondor.apana.org.au> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
participants (4)
-
chenxiang -
chenxiang (M)
-
Corentin Labbe -
Herbert Xu