| Message ID | 20250320091446.3647918-3-quic_lxu5@quicinc.com (mailing list archive) |
|---|---|
| State | Handled Elsewhere |
| Headers | show |
| Series | Add support for gpdsp remoteproc on sa8775p | expand |
On 3/20/2025 2:44 PM, Ling Xu wrote: > The fastrpc driver has support for 5 types of remoteprocs. There are > some products which support GPDSP remoteprocs. Add changes to support > GPDSP remoteprocs. > > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> > Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> > --- > drivers/misc/fastrpc.c | 10 ++++++++-- > 1 file changed, 8 insertions(+), 2 deletions(-) > > diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c > index 7b7a22c91fe4..80aa554b3042 100644 > --- a/drivers/misc/fastrpc.c > +++ b/drivers/misc/fastrpc.c > @@ -28,7 +28,9 @@ > #define SDSP_DOMAIN_ID (2) > #define CDSP_DOMAIN_ID (3) > #define CDSP1_DOMAIN_ID (4) > -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ > +#define GDSP0_DOMAIN_ID (5) > +#define GDSP1_DOMAIN_ID (6) > +#define FASTRPC_DEV_MAX 7 /* adsp, mdsp, slpi, cdsp, cdsp1, gdsp0, gdsp1 */ > #define FASTRPC_MAX_SESSIONS 14 > #define FASTRPC_MAX_VMIDS 16 > #define FASTRPC_ALIGN 128 > @@ -107,7 +109,9 @@ > #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) > > static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", > - "sdsp", "cdsp", "cdsp1" }; > + "sdsp", "cdsp", > + "cdsp1", "gdsp0", > + "gdsp1" }; > struct fastrpc_phy_page { > u64 addr; /* physical address */ > u64 size; /* size of contiguous region */ > @@ -2338,6 +2342,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) > break; > case CDSP_DOMAIN_ID: > case CDSP1_DOMAIN_ID: > + case GDSP0_DOMAIN_ID: > + case GDSP1_DOMAIN_ID: Reviewed-by: Ekansh Gupta <ekansh.gupta@oss.qualcomm.com> Thanks, Ekansh > data->unsigned_support = true; > /* Create both device nodes so that we can allow both Signed and Unsigned PD */ > err = fastrpc_device_register(rdev, data, true, domains[domain_id]);
On 3/20/25 10:14 AM, Ling Xu wrote: > The fastrpc driver has support for 5 types of remoteprocs. There are > some products which support GPDSP remoteprocs. Add changes to support > GPDSP remoteprocs. > > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> > Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> > --- > drivers/misc/fastrpc.c | 10 ++++++++-- > 1 file changed, 8 insertions(+), 2 deletions(-) > > diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c > index 7b7a22c91fe4..80aa554b3042 100644 > --- a/drivers/misc/fastrpc.c > +++ b/drivers/misc/fastrpc.c > @@ -28,7 +28,9 @@ > #define SDSP_DOMAIN_ID (2) > #define CDSP_DOMAIN_ID (3) > #define CDSP1_DOMAIN_ID (4) > -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ > +#define GDSP0_DOMAIN_ID (5) > +#define GDSP1_DOMAIN_ID (6) > +#define FASTRPC_DEV_MAX 7 /* adsp, mdsp, slpi, cdsp, cdsp1, gdsp0, gdsp1 */ > #define FASTRPC_MAX_SESSIONS 14 > #define FASTRPC_MAX_VMIDS 16 > #define FASTRPC_ALIGN 128 > @@ -107,7 +109,9 @@ > #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) > > static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", > - "sdsp", "cdsp", "cdsp1" }; > + "sdsp", "cdsp", > + "cdsp1", "gdsp0", > + "gdsp1" }; > struct fastrpc_phy_page { > u64 addr; /* physical address */ > u64 size; /* size of contiguous region */ > @@ -2338,6 +2342,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) > break; > case CDSP_DOMAIN_ID: > case CDSP1_DOMAIN_ID: > + case GDSP0_DOMAIN_ID: > + case GDSP1_DOMAIN_ID: > data->unsigned_support = true; There's a comment above this hunk that is no longer valid: '/* Unsigned PD offloading is only supported on CDSP and CDSP1 */' I would say it can be removed altogether I would also support renaming "unsigned_support" which is very generic to something like allow_unsigned_pds Konrad
On 20/03/2025 09:14, Ling Xu wrote: > The fastrpc driver has support for 5 types of remoteprocs. There are > some products which support GPDSP remoteprocs. Add changes to support > GPDSP remoteprocs. > > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> > Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> > --- > drivers/misc/fastrpc.c | 10 ++++++++-- > 1 file changed, 8 insertions(+), 2 deletions(-) > > diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c > index 7b7a22c91fe4..80aa554b3042 100644 > --- a/drivers/misc/fastrpc.c > +++ b/drivers/misc/fastrpc.c > @@ -28,7 +28,9 @@ > #define SDSP_DOMAIN_ID (2) > #define CDSP_DOMAIN_ID (3) > #define CDSP1_DOMAIN_ID (4) > -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ > +#define GDSP0_DOMAIN_ID (5) > +#define GDSP1_DOMAIN_ID (6) We have already made the driver look silly here, Lets not add domain ids for each instance, which is not a scalable. Domain ids are strictly for a domain not each instance. > +#define FASTRPC_DEV_MAX 7 /* adsp, mdsp, slpi, cdsp, cdsp1, gdsp0, gdsp1 */ > #define FASTRPC_MAX_SESSIONS 14 > #define FASTRPC_MAX_VMIDS 16 > #define FASTRPC_ALIGN 128 > @@ -107,7 +109,9 @@ > #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) > > static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", > - "sdsp", "cdsp", "cdsp1" }; > + "sdsp", "cdsp", > + "cdsp1", "gdsp0", > + "gdsp1" }; > struct fastrpc_phy_page { > u64 addr; /* physical address */ > u64 size; /* size of contiguous region */ > @@ -2338,6 +2342,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) > break; > case CDSP_DOMAIN_ID: > case CDSP1_DOMAIN_ID: > + case GDSP0_DOMAIN_ID: > + case GDSP1_DOMAIN_ID: > data->unsigned_support = true; > /* Create both device nodes so that we can allow both Signed and Unsigned PD */ > err = fastrpc_device_register(rdev, data, true, domains[domain_id]); Can you try this patch: only compile tested. ---------------------------------->cut<--------------------------------------- From 3f8607557162e16673b26fa253d11cafdc4444cf Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> Date: Thu, 20 Mar 2025 17:07:05 +0000 Subject: [PATCH] misc: fastrpc: cleanup the domain names Currently the domain ids are added for each instance of domain, this is totally not scalable approch. Clean this mess and create domain ids for only domains not its instances. This patch also moves the domain ids to uapi header as this is required for FASTRPC_IOCTL_GET_DSP_INFO ioctl. Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> --- drivers/misc/fastrpc.c | 45 ++++++++++++++++++++----------------- include/uapi/misc/fastrpc.h | 7 ++++++ 2 files changed, 32 insertions(+), 20 deletions(-) diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c index 7b7a22c91fe4..b3932897a437 100644 --- a/drivers/misc/fastrpc.c +++ b/drivers/misc/fastrpc.c @@ -23,12 +23,6 @@ #include <uapi/misc/fastrpc.h> #include <linux/of_reserved_mem.h> -#define ADSP_DOMAIN_ID (0) -#define MDSP_DOMAIN_ID (1) -#define SDSP_DOMAIN_ID (2) -#define CDSP_DOMAIN_ID (3) -#define CDSP1_DOMAIN_ID (4) -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ #define FASTRPC_MAX_SESSIONS 14 #define FASTRPC_MAX_VMIDS 16 #define FASTRPC_ALIGN 128 @@ -106,8 +100,6 @@ #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) -static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", - "sdsp", "cdsp", "cdsp1" }; struct fastrpc_phy_page { u64 addr; /* physical address */ u64 size; /* size of contiguous region */ @@ -1769,7 +1761,7 @@ static int fastrpc_get_dsp_info(struct fastrpc_user *fl, char __user *argp) return -EFAULT; cap.capability = 0; - if (cap.domain >= FASTRPC_DEV_MAX) { + if (cap.domain >= FASTRPC_DOMAIN_MAX) { dev_err(&fl->cctx->rpdev->dev, "Error: Invalid domain id:%d, err:%d\n", cap.domain, err); return -ECHRNG; @@ -2255,6 +2247,24 @@ static int fastrpc_device_register(struct device *dev, struct fastrpc_channel_ct return err; } +static int fastrpc_get_domain_id(const char *domain) +{ + if (strncmp(domain, "adsp", 4) == 0) { + return ADSP_DOMAIN_ID; + } else if (strncmp(domain, "cdsp", 4) == 0) { + return CDSP_DOMAIN_ID; + } else if (strncmp(domain, "mdsp", 4) ==0) { + return MDSP_DOMAIN_ID; + } else if (strncmp(domain, "sdsp", 4) ==0) { + return SDSP_DOMAIN_ID; + } else if (strncmp(domain, "gdsp", 4) ==0) { + return GDSP_DOMAIN_ID; + } + + return -EINVAL; + +} + static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) { struct device *rdev = &rpdev->dev; @@ -2272,15 +2282,10 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) return err; } - for (i = 0; i < FASTRPC_DEV_MAX; i++) { - if (!strcmp(domains[i], domain)) { - domain_id = i; - break; - } - } + domain_id = fastrpc_get_domain_id(domain); if (domain_id < 0) { - dev_info(rdev, "FastRPC Invalid Domain ID %d\n", domain_id); + dev_info(rdev, "FastRPC Domain %s not supported\n", domain); return -EINVAL; } @@ -2332,19 +2337,19 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) case SDSP_DOMAIN_ID: /* Unsigned PD offloading is only supported on CDSP and CDSP1 */ data->unsigned_support = false; - err = fastrpc_device_register(rdev, data, secure_dsp, domains[domain_id]); + err = fastrpc_device_register(rdev, data, secure_dsp, domain); if (err) goto fdev_error; break; case CDSP_DOMAIN_ID: - case CDSP1_DOMAIN_ID: + case GDSP_DOMAIN_ID: data->unsigned_support = true; /* Create both device nodes so that we can allow both Signed and Unsigned PD */ - err = fastrpc_device_register(rdev, data, true, domains[domain_id]); + err = fastrpc_device_register(rdev, data, true, domain); if (err) goto fdev_error; - err = fastrpc_device_register(rdev, data, false, domains[domain_id]); + err = fastrpc_device_register(rdev, data, false, domain); if (err) goto populate_error; break; diff --git a/include/uapi/misc/fastrpc.h b/include/uapi/misc/fastrpc.h index f33d914d8f46..89516abd258f 100644 --- a/include/uapi/misc/fastrpc.h +++ b/include/uapi/misc/fastrpc.h @@ -133,6 +133,13 @@ struct fastrpc_mem_unmap { __s32 reserved[5]; }; +#define ADSP_DOMAIN_ID (0) +#define MDSP_DOMAIN_ID (1) +#define SDSP_DOMAIN_ID (2) +#define CDSP_DOMAIN_ID (3) +#define GDSP_DOMAIN_ID (4) + +#define FASTRPC_DOMAIN_MAX 4 struct fastrpc_ioctl_capability { __u32 domain; __u32 attribute_id;
On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: > > > On 20/03/2025 09:14, Ling Xu wrote: > > The fastrpc driver has support for 5 types of remoteprocs. There are > > some products which support GPDSP remoteprocs. Add changes to support > > GPDSP remoteprocs. > > > > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> > > Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> > > --- > > drivers/misc/fastrpc.c | 10 ++++++++-- > > 1 file changed, 8 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c > > index 7b7a22c91fe4..80aa554b3042 100644 > > --- a/drivers/misc/fastrpc.c > > +++ b/drivers/misc/fastrpc.c > > @@ -28,7 +28,9 @@ > > #define SDSP_DOMAIN_ID (2) > > #define CDSP_DOMAIN_ID (3) > > #define CDSP1_DOMAIN_ID (4) > > -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ > > +#define GDSP0_DOMAIN_ID (5) > > +#define GDSP1_DOMAIN_ID (6) > > We have already made the driver look silly here, Lets not add domain ids for > each instance, which is not a scalable. > > Domain ids are strictly for a domain not each instance. Then CDSP1 should also be gone, correct?
On 20/03/2025 18:43, Dmitry Baryshkov wrote: > On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: >> >> >> On 20/03/2025 09:14, Ling Xu wrote: >>> The fastrpc driver has support for 5 types of remoteprocs. There are >>> some products which support GPDSP remoteprocs. Add changes to support >>> GPDSP remoteprocs. >>> >>> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> >>> Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> >>> --- >>> drivers/misc/fastrpc.c | 10 ++++++++-- >>> 1 file changed, 8 insertions(+), 2 deletions(-) >>> >>> diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c >>> index 7b7a22c91fe4..80aa554b3042 100644 >>> --- a/drivers/misc/fastrpc.c >>> +++ b/drivers/misc/fastrpc.c >>> @@ -28,7 +28,9 @@ >>> #define SDSP_DOMAIN_ID (2) >>> #define CDSP_DOMAIN_ID (3) >>> #define CDSP1_DOMAIN_ID (4) >>> -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ >>> +#define GDSP0_DOMAIN_ID (5) >>> +#define GDSP1_DOMAIN_ID (6) >> >> We have already made the driver look silly here, Lets not add domain ids for >> each instance, which is not a scalable. >> >> Domain ids are strictly for a domain not each instance. > > Then CDSP1 should also be gone, correct? Its already gone as part of the patch that I shared in this discussion. I will send a proper patch to list once Ling/Ekansh has agree with it. --srini >
On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: > > > On 20/03/2025 09:14, Ling Xu wrote: > > The fastrpc driver has support for 5 types of remoteprocs. There are > > some products which support GPDSP remoteprocs. Add changes to support > > GPDSP remoteprocs. > > > > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> > > Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> > > --- > > drivers/misc/fastrpc.c | 10 ++++++++-- > > 1 file changed, 8 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c > > index 7b7a22c91fe4..80aa554b3042 100644 > > --- a/drivers/misc/fastrpc.c > > +++ b/drivers/misc/fastrpc.c > > @@ -28,7 +28,9 @@ > > #define SDSP_DOMAIN_ID (2) > > #define CDSP_DOMAIN_ID (3) > > #define CDSP1_DOMAIN_ID (4) > > -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ > > +#define GDSP0_DOMAIN_ID (5) > > +#define GDSP1_DOMAIN_ID (6) > > We have already made the driver look silly here, Lets not add domain ids for > each instance, which is not a scalable. > > Domain ids are strictly for a domain not each instance. > > > > +#define FASTRPC_DEV_MAX 7 /* adsp, mdsp, slpi, cdsp, cdsp1, gdsp0, gdsp1 */ > > #define FASTRPC_MAX_SESSIONS 14 > > #define FASTRPC_MAX_VMIDS 16 > > #define FASTRPC_ALIGN 128 > > @@ -107,7 +109,9 @@ > > #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) > > static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", > > - "sdsp", "cdsp", "cdsp1" }; > > + "sdsp", "cdsp", > > + "cdsp1", "gdsp0", > > + "gdsp1" }; > > struct fastrpc_phy_page { > > u64 addr; /* physical address */ > > u64 size; /* size of contiguous region */ > > @@ -2338,6 +2342,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) > > break; > > case CDSP_DOMAIN_ID: > > case CDSP1_DOMAIN_ID: > > + case GDSP0_DOMAIN_ID: > > + case GDSP1_DOMAIN_ID: > > data->unsigned_support = true; > > /* Create both device nodes so that we can allow both Signed and Unsigned PD */ > > err = fastrpc_device_register(rdev, data, true, domains[domain_id]); > > > Can you try this patch: only compile tested. > > ---------------------------------->cut<--------------------------------------- > From 3f8607557162e16673b26fa253d11cafdc4444cf Mon Sep 17 00:00:00 2001 > From: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> > Date: Thu, 20 Mar 2025 17:07:05 +0000 > Subject: [PATCH] misc: fastrpc: cleanup the domain names > > Currently the domain ids are added for each instance of domain, this is > totally not scalable approch. > > Clean this mess and create domain ids for only domains not its > instances. > This patch also moves the domain ids to uapi header as this is required > for FASTRPC_IOCTL_GET_DSP_INFO ioctl. > > Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> > --- > drivers/misc/fastrpc.c | 45 ++++++++++++++++++++----------------- > include/uapi/misc/fastrpc.h | 7 ++++++ > 2 files changed, 32 insertions(+), 20 deletions(-) > > diff --git a/include/uapi/misc/fastrpc.h b/include/uapi/misc/fastrpc.h > index f33d914d8f46..89516abd258f 100644 > --- a/include/uapi/misc/fastrpc.h > +++ b/include/uapi/misc/fastrpc.h > @@ -133,6 +133,13 @@ struct fastrpc_mem_unmap { > __s32 reserved[5]; > }; > > +#define ADSP_DOMAIN_ID (0) > +#define MDSP_DOMAIN_ID (1) > +#define SDSP_DOMAIN_ID (2) > +#define CDSP_DOMAIN_ID (3) > +#define GDSP_DOMAIN_ID (4) Why are you adding these to uAPI? How are they going to be used by the userspace? > + > +#define FASTRPC_DOMAIN_MAX 4 > struct fastrpc_ioctl_capability { > __u32 domain; > __u32 attribute_id; > -- > 2.25.1 > > > ---------------------------------->cut<---------------------------------------
On Fri, Mar 21, 2025 at 12:23:15PM +0000, Srinivas Kandagatla wrote: > > > On 20/03/2025 18:43, Dmitry Baryshkov wrote: > > On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: > > > > > > > > > On 20/03/2025 09:14, Ling Xu wrote: > > > > The fastrpc driver has support for 5 types of remoteprocs. There are > > > > some products which support GPDSP remoteprocs. Add changes to support > > > > GPDSP remoteprocs. > > > > > > > > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> > > > > Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> > > > > --- > > > > drivers/misc/fastrpc.c | 10 ++++++++-- > > > > 1 file changed, 8 insertions(+), 2 deletions(-) > > > > > > > > diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c > > > > index 7b7a22c91fe4..80aa554b3042 100644 > > > > --- a/drivers/misc/fastrpc.c > > > > +++ b/drivers/misc/fastrpc.c > > > > @@ -28,7 +28,9 @@ > > > > #define SDSP_DOMAIN_ID (2) > > > > #define CDSP_DOMAIN_ID (3) > > > > #define CDSP1_DOMAIN_ID (4) > > > > -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ > > > > +#define GDSP0_DOMAIN_ID (5) > > > > +#define GDSP1_DOMAIN_ID (6) > > > > > > We have already made the driver look silly here, Lets not add domain ids for > > > each instance, which is not a scalable. > > > > > > Domain ids are strictly for a domain not each instance. > > > > Then CDSP1 should also be gone, correct? > Its already gone as part of the patch that I shared in this discussion. > > I will send a proper patch to list once Ling/Ekansh has agree with it. I see. For some reason mobile gmail decided that your patch is a quoted message and ignored it.
On 21/03/2025 14:07, Dmitry Baryshkov wrote: > On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: >> >> >> On 20/03/2025 09:14, Ling Xu wrote: >>> The fastrpc driver has support for 5 types of remoteprocs. There are >>> some products which support GPDSP remoteprocs. Add changes to support >>> GPDSP remoteprocs. >>> >>> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> >>> Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> >>> --- >>> drivers/misc/fastrpc.c | 10 ++++++++-- >>> 1 file changed, 8 insertions(+), 2 deletions(-) >>> >>> diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c >>> index 7b7a22c91fe4..80aa554b3042 100644 >>> --- a/drivers/misc/fastrpc.c >>> +++ b/drivers/misc/fastrpc.c >>> @@ -28,7 +28,9 @@ >>> #define SDSP_DOMAIN_ID (2) >>> #define CDSP_DOMAIN_ID (3) >>> #define CDSP1_DOMAIN_ID (4) >>> -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ >>> +#define GDSP0_DOMAIN_ID (5) >>> +#define GDSP1_DOMAIN_ID (6) >> >> We have already made the driver look silly here, Lets not add domain ids for >> each instance, which is not a scalable. >> >> Domain ids are strictly for a domain not each instance. >> >> >>> +#define FASTRPC_DEV_MAX 7 /* adsp, mdsp, slpi, cdsp, cdsp1, gdsp0, gdsp1 */ >>> #define FASTRPC_MAX_SESSIONS 14 >>> #define FASTRPC_MAX_VMIDS 16 >>> #define FASTRPC_ALIGN 128 >>> @@ -107,7 +109,9 @@ >>> #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) >>> static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", >>> - "sdsp", "cdsp", "cdsp1" }; >>> + "sdsp", "cdsp", >>> + "cdsp1", "gdsp0", >>> + "gdsp1" }; >>> struct fastrpc_phy_page { >>> u64 addr; /* physical address */ >>> u64 size; /* size of contiguous region */ >>> @@ -2338,6 +2342,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) >>> break; >>> case CDSP_DOMAIN_ID: >>> case CDSP1_DOMAIN_ID: >>> + case GDSP0_DOMAIN_ID: >>> + case GDSP1_DOMAIN_ID: >>> data->unsigned_support = true; >>> /* Create both device nodes so that we can allow both Signed and Unsigned PD */ >>> err = fastrpc_device_register(rdev, data, true, domains[domain_id]); >> >> >> Can you try this patch: only compile tested. >> >> ---------------------------------->cut<--------------------------------------- >> From 3f8607557162e16673b26fa253d11cafdc4444cf Mon Sep 17 00:00:00 2001 >> From: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> >> Date: Thu, 20 Mar 2025 17:07:05 +0000 >> Subject: [PATCH] misc: fastrpc: cleanup the domain names >> >> Currently the domain ids are added for each instance of domain, this is >> totally not scalable approch. >> >> Clean this mess and create domain ids for only domains not its >> instances. >> This patch also moves the domain ids to uapi header as this is required >> for FASTRPC_IOCTL_GET_DSP_INFO ioctl. >> >> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> >> --- >> drivers/misc/fastrpc.c | 45 ++++++++++++++++++++----------------- >> include/uapi/misc/fastrpc.h | 7 ++++++ >> 2 files changed, 32 insertions(+), 20 deletions(-) >> > > >> diff --git a/include/uapi/misc/fastrpc.h b/include/uapi/misc/fastrpc.h >> index f33d914d8f46..89516abd258f 100644 >> --- a/include/uapi/misc/fastrpc.h >> +++ b/include/uapi/misc/fastrpc.h >> @@ -133,6 +133,13 @@ struct fastrpc_mem_unmap { >> __s32 reserved[5]; >> }; >> >> +#define ADSP_DOMAIN_ID (0) >> +#define MDSP_DOMAIN_ID (1) >> +#define SDSP_DOMAIN_ID (2) >> +#define CDSP_DOMAIN_ID (3) >> +#define GDSP_DOMAIN_ID (4) > > Why are you adding these to uAPI? How are they going to be used by the > userspace? > >> + >> +#define FASTRPC_DOMAIN_MAX 4 >> struct fastrpc_ioctl_capability { >> __u32 domain; here, in domain value of fastrpc_ioctl_capability. >> __u32 attribute_id; >> -- >> 2.25.1 >> >> >> ---------------------------------->cut<--------------------------------------- >
On Mon, Mar 24, 2025 at 01:29:30PM +0000, Srinivas Kandagatla wrote: > > > On 21/03/2025 14:07, Dmitry Baryshkov wrote: > > On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: > > > > > > > > > On 20/03/2025 09:14, Ling Xu wrote: > > > > The fastrpc driver has support for 5 types of remoteprocs. There are > > > > some products which support GPDSP remoteprocs. Add changes to support > > > > GPDSP remoteprocs. > > > > > > > > Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> > > > > Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> > > > > --- > > > > drivers/misc/fastrpc.c | 10 ++++++++-- > > > > 1 file changed, 8 insertions(+), 2 deletions(-) > > > > > > > > diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c > > > > index 7b7a22c91fe4..80aa554b3042 100644 > > > > --- a/drivers/misc/fastrpc.c > > > > +++ b/drivers/misc/fastrpc.c > > > > @@ -28,7 +28,9 @@ > > > > #define SDSP_DOMAIN_ID (2) > > > > #define CDSP_DOMAIN_ID (3) > > > > #define CDSP1_DOMAIN_ID (4) > > > > -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ > > > > +#define GDSP0_DOMAIN_ID (5) > > > > +#define GDSP1_DOMAIN_ID (6) > > > > > > We have already made the driver look silly here, Lets not add domain ids for > > > each instance, which is not a scalable. > > > > > > Domain ids are strictly for a domain not each instance. > > > > > > > > > > +#define FASTRPC_DEV_MAX 7 /* adsp, mdsp, slpi, cdsp, cdsp1, gdsp0, gdsp1 */ > > > > #define FASTRPC_MAX_SESSIONS 14 > > > > #define FASTRPC_MAX_VMIDS 16 > > > > #define FASTRPC_ALIGN 128 > > > > @@ -107,7 +109,9 @@ > > > > #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) > > > > static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", > > > > - "sdsp", "cdsp", "cdsp1" }; > > > > + "sdsp", "cdsp", > > > > + "cdsp1", "gdsp0", > > > > + "gdsp1" }; > > > > struct fastrpc_phy_page { > > > > u64 addr; /* physical address */ > > > > u64 size; /* size of contiguous region */ > > > > @@ -2338,6 +2342,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) > > > > break; > > > > case CDSP_DOMAIN_ID: > > > > case CDSP1_DOMAIN_ID: > > > > + case GDSP0_DOMAIN_ID: > > > > + case GDSP1_DOMAIN_ID: > > > > data->unsigned_support = true; > > > > /* Create both device nodes so that we can allow both Signed and Unsigned PD */ > > > > err = fastrpc_device_register(rdev, data, true, domains[domain_id]); > > > > > > > > > Can you try this patch: only compile tested. > > > > > > ---------------------------------->cut<--------------------------------------- > > > From 3f8607557162e16673b26fa253d11cafdc4444cf Mon Sep 17 00:00:00 2001 > > > From: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> > > > Date: Thu, 20 Mar 2025 17:07:05 +0000 > > > Subject: [PATCH] misc: fastrpc: cleanup the domain names > > > > > > Currently the domain ids are added for each instance of domain, this is > > > totally not scalable approch. > > > > > > Clean this mess and create domain ids for only domains not its > > > instances. > > > This patch also moves the domain ids to uapi header as this is required > > > for FASTRPC_IOCTL_GET_DSP_INFO ioctl. > > > > > > Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org> > > > --- > > > drivers/misc/fastrpc.c | 45 ++++++++++++++++++++----------------- > > > include/uapi/misc/fastrpc.h | 7 ++++++ > > > 2 files changed, 32 insertions(+), 20 deletions(-) > > > > > > > > > > diff --git a/include/uapi/misc/fastrpc.h b/include/uapi/misc/fastrpc.h > > > index f33d914d8f46..89516abd258f 100644 > > > --- a/include/uapi/misc/fastrpc.h > > > +++ b/include/uapi/misc/fastrpc.h > > > @@ -133,6 +133,13 @@ struct fastrpc_mem_unmap { > > > __s32 reserved[5]; > > > }; > > > > > > +#define ADSP_DOMAIN_ID (0) > > > +#define MDSP_DOMAIN_ID (1) > > > +#define SDSP_DOMAIN_ID (2) > > > +#define CDSP_DOMAIN_ID (3) > > > +#define GDSP_DOMAIN_ID (4) > > > > Why are you adding these to uAPI? How are they going to be used by the > > userspace? > > > > > + > > > +#define FASTRPC_DOMAIN_MAX 4 > > > struct fastrpc_ioctl_capability { > > > __u32 domain; > > here, in domain value of fastrpc_ioctl_capability. I see. I'd say, moving those IDs to a header is a separate change. Also please document that this patch changes the user-visible value for CDSP1 DSP. > > > > > > __u32 attribute_id; > > > -- > > > 2.25.1 > > > > > > > > > ---------------------------------->cut<--------------------------------------- > >
On 3/21/2025 5:53 PM, Srinivas Kandagatla wrote: > > > On 20/03/2025 18:43, Dmitry Baryshkov wrote: >> On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: >>> >>> >>> On 20/03/2025 09:14, Ling Xu wrote: >>>> The fastrpc driver has support for 5 types of remoteprocs. There are >>>> some products which support GPDSP remoteprocs. Add changes to support >>>> GPDSP remoteprocs. >>>> >>>> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> >>>> Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> >>>> --- >>>> drivers/misc/fastrpc.c | 10 ++++++++-- >>>> 1 file changed, 8 insertions(+), 2 deletions(-) >>>> >>>> diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c >>>> index 7b7a22c91fe4..80aa554b3042 100644 >>>> --- a/drivers/misc/fastrpc.c >>>> +++ b/drivers/misc/fastrpc.c >>>> @@ -28,7 +28,9 @@ >>>> #define SDSP_DOMAIN_ID (2) >>>> #define CDSP_DOMAIN_ID (3) >>>> #define CDSP1_DOMAIN_ID (4) >>>> -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ >>>> +#define GDSP0_DOMAIN_ID (5) >>>> +#define GDSP1_DOMAIN_ID (6) >>> >>> We have already made the driver look silly here, Lets not add domain ids for >>> each instance, which is not a scalable. >>> >>> Domain ids are strictly for a domain not each instance. >> >> Then CDSP1 should also be gone, correct? > Its already gone as part of the patch that I shared in this discussion. > > I will send a proper patch to list once Ling/Ekansh has agree with it. > Thanks, Srini, for sharing this clean-up patch. It looks proper to me, but I was thinking if we could remove the domain_id dependency from the fastrpc driver. The addition of any new DSP will frequently require changes in the driver. Currently, its usage is for creating different types of device nodes and transferring memory ownership to SLPI when a memory region is added. The actual intention behind different types of device nodes can be defined as follows: fastrpc-xdsp-secure: Used for signed (privileged) PD offload and for daemons. fastrpc-xdsp: Should be used only for unsigned (less privileged) PD offload. The reason for this constraint is to prevent any untrusted process from communicating with any privileged PD on DSP, which poses a security risk. The access to different device nodes can be provided/restricted based on UID/GID (still need to check more on this; on Android-like systems, this is controlled by SELinux). There is already a qcom,non-secure-domain device tree property[1] which doesn't have a proper definition as of today. The actual way to differentiate between secure and non-secure DSP should be based on its ability to support unsigned PD. One way to remove the domain_id dependency that I can think of is to use this property to create different types of device nodes. Essentially, if unsigned PD is supported (e.g., CDSP, GPDSP), we add this property to the DT node and create both types of device nodes based on this. Otherwise, only the secure device node is created. This raises the question of backward compatibility, but I see that on most older platform DTs, this property is already added, so both device nodes will be created there, and applications will work as expected. If any old DT DSP node lacks this property, we can add it there as well. Going forward, the qcom-non-secure-property should be added only if unsigned PD is supported. This way, we can clean up the driver completely to remove the domain_id dependency. If this sounds good, I can work on this design and send out a patch. [1] https://web.git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/devicetree/bindings/misc/qcom,fastrpc.yaml#n44 --Ekansh > --srini >> >
On 02/04/2025 11:38, Ekansh Gupta wrote: > > > On 3/21/2025 5:53 PM, Srinivas Kandagatla wrote: >> >> >> On 20/03/2025 18:43, Dmitry Baryshkov wrote: >>> On Thu, Mar 20, 2025 at 05:11:20PM +0000, Srinivas Kandagatla wrote: >>>> >>>> >>>> On 20/03/2025 09:14, Ling Xu wrote: >>>>> The fastrpc driver has support for 5 types of remoteprocs. There are >>>>> some products which support GPDSP remoteprocs. Add changes to support >>>>> GPDSP remoteprocs. >>>>> >>>>> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> >>>>> Signed-off-by: Ling Xu <quic_lxu5@quicinc.com> >>>>> --- >>>>> drivers/misc/fastrpc.c | 10 ++++++++-- >>>>> 1 file changed, 8 insertions(+), 2 deletions(-) >>>>> >>>>> diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c >>>>> index 7b7a22c91fe4..80aa554b3042 100644 >>>>> --- a/drivers/misc/fastrpc.c >>>>> +++ b/drivers/misc/fastrpc.c >>>>> @@ -28,7 +28,9 @@ >>>>> #define SDSP_DOMAIN_ID (2) >>>>> #define CDSP_DOMAIN_ID (3) >>>>> #define CDSP1_DOMAIN_ID (4) >>>>> -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ >>>>> +#define GDSP0_DOMAIN_ID (5) >>>>> +#define GDSP1_DOMAIN_ID (6) >>>> >>>> We have already made the driver look silly here, Lets not add domain ids for >>>> each instance, which is not a scalable. >>>> >>>> Domain ids are strictly for a domain not each instance. >>> >>> Then CDSP1 should also be gone, correct? >> Its already gone as part of the patch that I shared in this discussion. >> >> I will send a proper patch to list once Ling/Ekansh has agree with it. >> > Thanks, Srini, for sharing this clean-up patch. It looks proper to > me, but I was thinking if we could remove the domain_id dependency > from the fastrpc driver. The addition of any new DSP will frequently > require changes in the driver. Currently, its usage is for creating > different types of device nodes and transferring memory ownership to > SLPI when a memory region is added. > > The actual intention behind different types of device nodes can be > defined as follows: > > fastrpc-xdsp-secure: Used for signed (privileged) PD offload and for daemons. > fastrpc-xdsp: Should be used only for unsigned (less privileged) PD offload. > > The reason for this constraint is to prevent any untrusted process > from communicating with any privileged PD on DSP, which poses a security risk. > The access to different device nodes can be provided/restricted based on UID/GID > (still need to check more on this; on Android-like systems, this is controlled by > SELinux). > > There is already a qcom,non-secure-domain device tree property[1] which doesn't > have a proper definition as of today. The actual way to differentiate between > secure and non-secure DSP should be based on its ability to support unsigned PD. > > One way to remove the domain_id dependency that I can think of is to use this > property to create different types of device nodes. Essentially, if unsigned PD > is supported (e.g., CDSP, GPDSP), we add this property to the DT node and create > both types of device nodes based on this. Otherwise, only the secure device node > is created. This sounds like breaking backwards compatibility on the userspace side. You can not do that. > > This raises the question of backward compatibility, but I see that on most older > platform DTs, this property is already added, so both device nodes will be created > there, and applications will work as expected. If any old DT DSP node lacks this > property, we can add it there as well. > > Going forward, the qcom-non-secure-property should be added only if unsigned PD > is supported. This way, we can clean up the driver completely to remove the > domain_id dependency. > > If this sounds good, I can work on this design and send out a patch. > > [1] https://web.git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/Documentation/devicetree/bindings/misc/qcom,fastrpc.yaml#n44 > > --Ekansh > >> --srini >>> >> >
diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c index 7b7a22c91fe4..80aa554b3042 100644 --- a/drivers/misc/fastrpc.c +++ b/drivers/misc/fastrpc.c @@ -28,7 +28,9 @@ #define SDSP_DOMAIN_ID (2) #define CDSP_DOMAIN_ID (3) #define CDSP1_DOMAIN_ID (4) -#define FASTRPC_DEV_MAX 5 /* adsp, mdsp, slpi, cdsp, cdsp1 */ +#define GDSP0_DOMAIN_ID (5) +#define GDSP1_DOMAIN_ID (6) +#define FASTRPC_DEV_MAX 7 /* adsp, mdsp, slpi, cdsp, cdsp1, gdsp0, gdsp1 */ #define FASTRPC_MAX_SESSIONS 14 #define FASTRPC_MAX_VMIDS 16 #define FASTRPC_ALIGN 128 @@ -107,7 +109,9 @@ #define miscdev_to_fdevice(d) container_of(d, struct fastrpc_device, miscdev) static const char *domains[FASTRPC_DEV_MAX] = { "adsp", "mdsp", - "sdsp", "cdsp", "cdsp1" }; + "sdsp", "cdsp", + "cdsp1", "gdsp0", + "gdsp1" }; struct fastrpc_phy_page { u64 addr; /* physical address */ u64 size; /* size of contiguous region */ @@ -2338,6 +2342,8 @@ static int fastrpc_rpmsg_probe(struct rpmsg_device *rpdev) break; case CDSP_DOMAIN_ID: case CDSP1_DOMAIN_ID: + case GDSP0_DOMAIN_ID: + case GDSP1_DOMAIN_ID: data->unsigned_support = true; /* Create both device nodes so that we can allow both Signed and Unsigned PD */ err = fastrpc_device_register(rdev, data, true, domains[domain_id]);