Commit c1e98807 authored by Lucas Segarra Fernandez's avatar Lucas Segarra Fernandez Committed by Herbert Xu

crypto: qat - extend buffer list logic interface

Extend qat_bl_sgl_to_bufl() to allow skipping the mapping of a region
of the source and the destination scatter lists starting from byte
zero.

This is to support the ZLIB format (RFC 1950) in the qat driver.
The ZLIB format is made of deflate compressed data surrounded by a
header and a footer. The QAT accelerators support only the deflate
algorithm, therefore the header should not be mapped since it is
inserted in software.
Signed-off-by: default avatarLucas Segarra Fernandez <lucas.segarra.fernandez@intel.com>
Reviewed-by: default avatarGiovanni Cabiddu <giovanni.cabiddu@intel.com>
Signed-off-by: default avatarHerbert Xu <herbert@gondor.apana.org.au>
parent d52b0c78
......@@ -53,6 +53,8 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
struct qat_request_buffs *buf,
dma_addr_t extra_dst_buff,
size_t sz_extra_dst_buff,
unsigned int sskip,
unsigned int dskip,
gfp_t flags)
{
struct device *dev = &GET_DEV(accel_dev);
......@@ -65,6 +67,7 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
struct scatterlist *sg;
size_t sz_out, sz = struct_size(bufl, buffers, n);
int node = dev_to_node(&GET_DEV(accel_dev));
unsigned int left;
int bufl_dma_dir;
if (unlikely(!n))
......@@ -88,19 +91,29 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
for (i = 0; i < n; i++)
bufl->buffers[i].addr = DMA_MAPPING_ERROR;
left = sskip;
for_each_sg(sgl, sg, n, i) {
int y = sg_nctr;
if (!sg->length)
continue;
bufl->buffers[y].addr = dma_map_single(dev, sg_virt(sg),
sg->length,
if (left >= sg->length) {
left -= sg->length;
continue;
}
bufl->buffers[y].addr = dma_map_single(dev, sg_virt(sg) + left,
sg->length - left,
bufl_dma_dir);
bufl->buffers[y].len = sg->length;
if (unlikely(dma_mapping_error(dev, bufl->buffers[y].addr)))
goto err_in;
sg_nctr++;
if (left) {
bufl->buffers[y].len -= left;
left = 0;
}
}
bufl->num_bufs = sg_nctr;
blp = dma_map_single(dev, bufl, sz, DMA_TO_DEVICE);
......@@ -117,6 +130,8 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
n = n_sglout + extra_buff;
sz_out = struct_size(buflout, buffers, n);
left = dskip;
sg_nctr = 0;
if (n > QAT_MAX_BUFF_DESC) {
......@@ -139,13 +154,21 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
if (!sg->length)
continue;
buffers[y].addr = dma_map_single(dev, sg_virt(sg),
sg->length,
if (left >= sg->length) {
left -= sg->length;
continue;
}
buffers[y].addr = dma_map_single(dev, sg_virt(sg) + left,
sg->length - left,
DMA_FROM_DEVICE);
if (unlikely(dma_mapping_error(dev, buffers[y].addr)))
goto err_out;
buffers[y].len = sg->length;
sg_nctr++;
if (left) {
buffers[y].len -= left;
left = 0;
}
}
if (extra_buff) {
buffers[sg_nctr].addr = extra_dst_buff;
......@@ -212,15 +235,19 @@ int qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev,
{
dma_addr_t extra_dst_buff = 0;
size_t sz_extra_dst_buff = 0;
unsigned int sskip = 0;
unsigned int dskip = 0;
if (params) {
extra_dst_buff = params->extra_dst_buff;
sz_extra_dst_buff = params->sz_extra_dst_buff;
sskip = params->sskip;
dskip = params->dskip;
}
return __qat_bl_sgl_to_bufl(accel_dev, sgl, sglout, buf,
extra_dst_buff, sz_extra_dst_buff,
flags);
sskip, dskip, flags);
}
static void qat_bl_sgl_unmap(struct adf_accel_dev *accel_dev,
......
......@@ -42,6 +42,8 @@ struct qat_request_buffs {
struct qat_sgl_to_bufl_params {
dma_addr_t extra_dst_buff;
size_t sz_extra_dst_buff;
unsigned int sskip;
unsigned int dskip;
};
void qat_bl_free_bufl(struct adf_accel_dev *accel_dev,
......
......@@ -233,6 +233,9 @@ static int qat_comp_alg_compress_decompress(struct acomp_req *areq,
size_t ovf_buff_sz;
int ret;
params.sskip = 0;
params.dskip = 0;
if (!areq->src || !slen)
return -EINVAL;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment