diff options
Diffstat (limited to 'drivers/gpu/host1x/hw')
| -rw-r--r-- | drivers/gpu/host1x/hw/cdma_hw.c | 34 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/channel_hw.c | 137 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x01_hardware.h | 114 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x02_hardware.h | 113 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x04_hardware.h | 113 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x05_hardware.h | 113 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x06_hardware.h | 118 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x07_hardware.h | 118 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x08.c | 33 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x08.h | 15 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/host1x08_hardware.h | 21 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/hw_host1x08_channel.h | 11 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/hw_host1x08_common.h | 11 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h | 9 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/hw_host1x08_uclass.h | 181 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/hw_host1x08_vm.h | 36 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/intr_hw.c | 11 | ||||
| -rw-r--r-- | drivers/gpu/host1x/hw/opcodes.h | 150 |
18 files changed, 622 insertions, 716 deletions
diff --git a/drivers/gpu/host1x/hw/cdma_hw.c b/drivers/gpu/host1x/hw/cdma_hw.c index e49cd5b8f735..1b65a10b9dfc 100644 --- a/drivers/gpu/host1x/hw/cdma_hw.c +++ b/drivers/gpu/host1x/hw/cdma_hw.c @@ -238,6 +238,37 @@ static void cdma_resume(struct host1x_cdma *cdma, u32 getptr) cdma_timeout_restart(cdma, getptr); } +static void timeout_release_mlock(struct host1x_cdma *cdma) +{ +#if HOST1X_HW >= 8 + /* Tegra186 and Tegra194 require a more complicated MLOCK release + * sequence. Furthermore, those chips by default don't enforce MLOCKs, + * so it turns out that if we don't /actually/ need MLOCKs, we can just + * ignore them. + * + * As such, for now just implement this on Tegra234 where things are + * stricter but also easy to implement. + */ + struct host1x_channel *ch = cdma_to_channel(cdma); + struct host1x *host1x = cdma_to_host1x(cdma); + u32 offset; + + switch (ch->client->class) { + case HOST1X_CLASS_VIC: + offset = HOST1X_COMMON_VIC_MLOCK; + break; + case HOST1X_CLASS_NVDEC: + offset = HOST1X_COMMON_NVDEC_MLOCK; + break; + default: + WARN(1, "%s was not updated for class %u", __func__, ch->client->class); + return; + } + + host1x_common_writel(host1x, 0x0, offset); +#endif +} + /* * If this timeout fires, it indicates the current sync_queue entry has * exceeded its TTL and the userctx should be timed out and remaining @@ -288,6 +319,9 @@ static void cdma_timeout_handler(struct work_struct *work) /* stop HW, resetting channel/module */ host1x_hw_cdma_freeze(host1x, cdma); + /* release any held MLOCK */ + timeout_release_mlock(cdma); + host1x_cdma_update_sync_queue(cdma, ch->dev); mutex_unlock(&cdma->lock); } diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c index 6b40e9af1e88..732abe0750ff 100644 --- a/drivers/gpu/host1x/hw/channel_hw.c +++ b/drivers/gpu/host1x/hw/channel_hw.c @@ -47,10 +47,41 @@ static void trace_write_gather(struct host1x_cdma *cdma, struct host1x_bo *bo, } } -static void submit_wait(struct host1x_cdma *cdma, u32 id, u32 threshold, +static void submit_wait(struct host1x_job *job, u32 id, u32 threshold, u32 next_class) { -#if HOST1X_HW >= 2 + struct host1x_cdma *cdma = &job->channel->cdma; + +#if HOST1X_HW >= 6 + u32 stream_id; + + /* + * If a memory context has been set, use it. Otherwise + * (if context isolation is disabled) use the engine's + * firmware stream ID. + */ + if (job->memory_context) + stream_id = job->memory_context->stream_id; + else + stream_id = job->engine_fallback_streamid; + + host1x_cdma_push_wide(cdma, + host1x_opcode_setclass( + HOST1X_CLASS_HOST1X, + HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32, + /* WAIT_SYNCPT_32 is at SYNCPT_PAYLOAD_32+2 */ + BIT(0) | BIT(2) + ), + threshold, + id, + HOST1X_OPCODE_NOP + ); + host1x_cdma_push_wide(&job->channel->cdma, + host1x_opcode_setclass(job->class, 0, 0), + host1x_opcode_setpayload(stream_id), + host1x_opcode_setstreamid(job->engine_streamid_offset / 4), + HOST1X_OPCODE_NOP); +#elif HOST1X_HW >= 2 host1x_cdma_push_wide(cdma, host1x_opcode_setclass( HOST1X_CLASS_HOST1X, @@ -97,7 +128,7 @@ static void submit_gathers(struct host1x_job *job, u32 job_syncpt_base) else threshold = cmd->wait.threshold; - submit_wait(cdma, cmd->wait.id, threshold, cmd->wait.next_class); + submit_wait(job, cmd->wait.id, threshold, cmd->wait.next_class); } else { struct host1x_job_gather *g = &cmd->gather; @@ -180,11 +211,77 @@ static void host1x_enable_gather_filter(struct host1x_channel *ch) #endif } +static void channel_program_cdma(struct host1x_job *job) +{ + struct host1x_cdma *cdma = &job->channel->cdma; + struct host1x_syncpt *sp = job->syncpt; + +#if HOST1X_HW >= 6 + u32 fence; + + /* Enter engine class with invalid stream ID. */ + host1x_cdma_push_wide(cdma, + host1x_opcode_acquire_mlock(job->class), + host1x_opcode_setclass(job->class, 0, 0), + host1x_opcode_setpayload(0), + host1x_opcode_setstreamid(job->engine_streamid_offset / 4)); + + /* Before switching stream ID to real stream ID, ensure engine is idle. */ + fence = host1x_syncpt_incr_max(sp, 1); + host1x_cdma_push(&job->channel->cdma, + host1x_opcode_nonincr(HOST1X_UCLASS_INCR_SYNCPT, 1), + HOST1X_UCLASS_INCR_SYNCPT_INDX_F(job->syncpt->id) | + HOST1X_UCLASS_INCR_SYNCPT_COND_F(4)); + submit_wait(job, job->syncpt->id, fence, job->class); + + /* Submit work. */ + job->syncpt_end = host1x_syncpt_incr_max(sp, job->syncpt_incrs); + submit_gathers(job, job->syncpt_end - job->syncpt_incrs); + + /* Before releasing MLOCK, ensure engine is idle again. */ + fence = host1x_syncpt_incr_max(sp, 1); + host1x_cdma_push(&job->channel->cdma, + host1x_opcode_nonincr(HOST1X_UCLASS_INCR_SYNCPT, 1), + HOST1X_UCLASS_INCR_SYNCPT_INDX_F(job->syncpt->id) | + HOST1X_UCLASS_INCR_SYNCPT_COND_F(4)); + submit_wait(job, job->syncpt->id, fence, job->class); + + /* Release MLOCK. */ + host1x_cdma_push(cdma, + HOST1X_OPCODE_NOP, host1x_opcode_release_mlock(job->class)); +#else + if (job->serialize) { + /* + * Force serialization by inserting a host wait for the + * previous job to finish before this one can commence. + */ + host1x_cdma_push(cdma, + host1x_opcode_setclass(HOST1X_CLASS_HOST1X, + host1x_uclass_wait_syncpt_r(), 1), + host1x_class_host_wait_syncpt(job->syncpt->id, + host1x_syncpt_read_max(sp))); + } + + /* Synchronize base register to allow using it for relative waiting */ + if (sp->base) + synchronize_syncpt_base(job); + + /* add a setclass for modules that require it */ + if (job->class) + host1x_cdma_push(cdma, + host1x_opcode_setclass(job->class, 0, 0), + HOST1X_OPCODE_NOP); + + job->syncpt_end = host1x_syncpt_incr_max(sp, job->syncpt_incrs); + + submit_gathers(job, job->syncpt_end - job->syncpt_incrs); +#endif +} + static int channel_submit(struct host1x_job *job) { struct host1x_channel *ch = job->channel; struct host1x_syncpt *sp = job->syncpt; - u32 user_syncpt_incrs = job->syncpt_incrs; u32 prev_max = 0; u32 syncval; int err; @@ -212,6 +309,7 @@ static int channel_submit(struct host1x_job *job) host1x_channel_set_streamid(ch); host1x_enable_gather_filter(ch); + host1x_hw_syncpt_assign_to_channel(host, sp, ch); /* begin a CDMA submit */ err = host1x_cdma_begin(&ch->cdma, job); @@ -220,35 +318,8 @@ static int channel_submit(struct host1x_job *job) goto error; } - if (job->serialize) { - /* - * Force serialization by inserting a host wait for the - * previous job to finish before this one can commence. - */ - host1x_cdma_push(&ch->cdma, - host1x_opcode_setclass(HOST1X_CLASS_HOST1X, - host1x_uclass_wait_syncpt_r(), 1), - host1x_class_host_wait_syncpt(job->syncpt->id, - host1x_syncpt_read_max(sp))); - } - - /* Synchronize base register to allow using it for relative waiting */ - if (sp->base) - synchronize_syncpt_base(job); - - syncval = host1x_syncpt_incr_max(sp, user_syncpt_incrs); - - host1x_hw_syncpt_assign_to_channel(host, sp, ch); - - job->syncpt_end = syncval; - - /* add a setclass for modules that require it */ - if (job->class) - host1x_cdma_push(&ch->cdma, - host1x_opcode_setclass(job->class, 0, 0), - HOST1X_OPCODE_NOP); - - submit_gathers(job, syncval - user_syncpt_incrs); + channel_program_cdma(job); + syncval = host1x_syncpt_read_max(sp); /* end CDMA submit & stash pinned hMems into sync queue */ host1x_cdma_end(&ch->cdma, job); diff --git a/drivers/gpu/host1x/hw/host1x01_hardware.h b/drivers/gpu/host1x/hw/host1x01_hardware.h index fe59df1d3dc3..cb93d7c1808c 100644 --- a/drivers/gpu/host1x/hw/host1x01_hardware.h +++ b/drivers/gpu/host1x/hw/host1x01_hardware.h @@ -15,118 +15,6 @@ #include "hw_host1x01_sync.h" #include "hw_host1x01_uclass.h" -static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h" #endif diff --git a/drivers/gpu/host1x/hw/host1x02_hardware.h b/drivers/gpu/host1x/hw/host1x02_hardware.h index af60d7fb016d..2d1282b9bc33 100644 --- a/drivers/gpu/host1x/hw/host1x02_hardware.h +++ b/drivers/gpu/host1x/hw/host1x02_hardware.h @@ -15,117 +15,6 @@ #include "hw_host1x02_sync.h" #include "hw_host1x02_uclass.h" -static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h" #endif diff --git a/drivers/gpu/host1x/hw/host1x04_hardware.h b/drivers/gpu/host1x/hw/host1x04_hardware.h index 4f9bcddf27e3..84d244e8af30 100644 --- a/drivers/gpu/host1x/hw/host1x04_hardware.h +++ b/drivers/gpu/host1x/hw/host1x04_hardware.h @@ -15,117 +15,6 @@ #include "hw_host1x04_sync.h" #include "hw_host1x04_uclass.h" -static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h" #endif diff --git a/drivers/gpu/host1x/hw/host1x05_hardware.h b/drivers/gpu/host1x/hw/host1x05_hardware.h index af3ab4b7f010..1dcde6ec7909 100644 --- a/drivers/gpu/host1x/hw/host1x05_hardware.h +++ b/drivers/gpu/host1x/hw/host1x05_hardware.h @@ -15,117 +15,6 @@ #include "hw_host1x05_sync.h" #include "hw_host1x05_uclass.h" -static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h" #endif diff --git a/drivers/gpu/host1x/hw/host1x06_hardware.h b/drivers/gpu/host1x/hw/host1x06_hardware.h index 01a142a09800..c05cfa7e3090 100644 --- a/drivers/gpu/host1x/hw/host1x06_hardware.h +++ b/drivers/gpu/host1x/hw/host1x06_hardware.h @@ -16,122 +16,6 @@ #include "hw_host1x06_vm.h" #include "hw_host1x06_hypervisor.h" -static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -static inline u32 host1x_opcode_gather_wide(unsigned count) -{ - return (12 << 28) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h" #endif diff --git a/drivers/gpu/host1x/hw/host1x07_hardware.h b/drivers/gpu/host1x/hw/host1x07_hardware.h index e6582172ebfd..d67364e03956 100644 --- a/drivers/gpu/host1x/hw/host1x07_hardware.h +++ b/drivers/gpu/host1x/hw/host1x07_hardware.h @@ -16,122 +16,6 @@ #include "hw_host1x07_vm.h" #include "hw_host1x07_hypervisor.h" -static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -static inline u32 host1x_opcode_gather_wide(unsigned count) -{ - return (12 << 28) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h" #endif diff --git a/drivers/gpu/host1x/hw/host1x08.c b/drivers/gpu/host1x/hw/host1x08.c new file mode 100644 index 000000000000..754890c34c74 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra234 SoCs + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x08.h" +#include "host1x08_hardware.h" + +/* include code */ +#define HOST1X_HW 8 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x08_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x08.h b/drivers/gpu/host1x/hw/host1x08.h new file mode 100644 index 000000000000..a6bad56e44cf --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra234 SoCs + * + * Copyright (c) 2018 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X08_H +#define HOST1X_HOST1X08_H + +struct host1x; + +int host1x08_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x08_hardware.h b/drivers/gpu/host1x/hw/host1x08_hardware.h new file mode 100644 index 000000000000..936243060bff --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08_hardware.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra234 + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X08_HARDWARE_H +#define __HOST1X_HOST1X08_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x08_uclass.h" +#include "hw_host1x08_vm.h" +#include "hw_host1x08_hypervisor.h" +#include "hw_host1x08_common.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_channel.h b/drivers/gpu/host1x/hw/hw_host1x08_channel.h new file mode 100644 index 000000000000..c9272d2ab14a --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_channel.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef HOST1X_HW_HOST1X08_CHANNEL_H +#define HOST1X_HW_HOST1X08_CHANNEL_H + +#define HOST1X_CHANNEL_SMMU_STREAMID 0x084 + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_common.h b/drivers/gpu/host1x/hw/hw_host1x08_common.h new file mode 100644 index 000000000000..8e0c99150ec2 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_common.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_COMMON_OFA_MLOCK 0x4050 +#define HOST1X_COMMON_NVJPG1_MLOCK 0x4070 +#define HOST1X_COMMON_VIC_MLOCK 0x4078 +#define HOST1X_COMMON_NVENC_MLOCK 0x407c +#define HOST1X_COMMON_NVDEC_MLOCK 0x4080 +#define HOST1X_COMMON_NVJPG_MLOCK 0x4084 diff --git a/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h b/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h new file mode 100644 index 000000000000..22964324c914 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h @@ -0,0 +1,9 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_HV_SYNCPT_PROT_EN 0x1724 +#define HOST1X_HV_SYNCPT_PROT_EN_CH_EN BIT(1) +#define HOST1X_HV_CH_MLOCK_EN(x) (0x1700 + (x * 4)) +#define HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(x) (0x1710 + (x * 4)) diff --git a/drivers/gpu/host1x/hw/hw_host1x08_uclass.h b/drivers/gpu/host1x/hw/hw_host1x08_uclass.h new file mode 100644 index 000000000000..724cccd71aa1 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X08_UCLASS_H +#define HOST1X_HW_HOST1X08_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 10; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_vm.h b/drivers/gpu/host1x/hw/hw_host1x08_vm.h new file mode 100644 index 000000000000..1455a4670bf8 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_vm.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_CHANNEL_DMASTART 0x0000 +#define HOST1X_CHANNEL_DMASTART_HI 0x0004 +#define HOST1X_CHANNEL_DMAPUT 0x0008 +#define HOST1X_CHANNEL_DMAPUT_HI 0x000c +#define HOST1X_CHANNEL_DMAGET 0x0010 +#define HOST1X_CHANNEL_DMAGET_HI 0x0014 +#define HOST1X_CHANNEL_DMAEND 0x0018 +#define HOST1X_CHANNEL_DMAEND_HI 0x001c +#define HOST1X_CHANNEL_DMACTRL 0x0020 +#define HOST1X_CHANNEL_DMACTRL_DMASTOP BIT(0) +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST BIT(1) +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET BIT(2) +#define HOST1X_CHANNEL_CMDFIFO_STAT 0x0024 +#define HOST1X_CHANNEL_CMDFIFO_STAT_EMPTY BIT(13) +#define HOST1X_CHANNEL_CMDFIFO_RDATA 0x0028 +#define HOST1X_CHANNEL_CMDP_OFFSET 0x0030 +#define HOST1X_CHANNEL_CMDP_CLASS 0x0034 +#define HOST1X_CHANNEL_CHANNELSTAT 0x0038 +#define HOST1X_CHANNEL_CMDPROC_STOP 0x0048 +#define HOST1X_CHANNEL_TEARDOWN 0x004c +#define HOST1X_CHANNEL_SMMU_STREAMID 0x0084 + +#define HOST1X_SYNC_SYNCPT_CPU_INCR(x) (0x6400 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(x) (0x6600 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INTR_DEST(x) (0x6684 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(x) (0x770c + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(x) (0x7790 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT(x) (0x8080 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INT_THRESH(x) (0xa088 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP(x) (0xb090 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP_CH(v) (((v) & 0x3f) << 8) diff --git a/drivers/gpu/host1x/hw/intr_hw.c b/drivers/gpu/host1x/hw/intr_hw.c index f56375ee6e71..9acccdb139e6 100644 --- a/drivers/gpu/host1x/hw/intr_hw.c +++ b/drivers/gpu/host1x/hw/intr_hw.c @@ -76,6 +76,17 @@ static void intr_hw_init(struct host1x *host, u32 cpm) /* update host clocks per usec */ host1x_sync_writel(host, cpm, HOST1X_SYNC_USEC_CLK); #endif +#if HOST1X_HW >= 8 + u32 id; + + /* + * Program threshold interrupt destination among 8 lines per VM, + * per syncpoint. For now, just direct all to the first interrupt + * line. + */ + for (id = 0; id < host->info->nb_pts; id++) + host1x_sync_writel(host, 0, HOST1X_SYNC_SYNCPT_INTR_DEST(id)); +#endif } static int diff --git a/drivers/gpu/host1x/hw/opcodes.h b/drivers/gpu/host1x/hw/opcodes.h new file mode 100644 index 000000000000..649614499b04 --- /dev/null +++ b/drivers/gpu/host1x/hw/opcodes.h @@ -0,0 +1,150 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x opcodes + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef __HOST1X_OPCODES_H +#define __HOST1X_OPCODES_H + +#include <linux/types.h> + +static inline u32 host1x_class_host_wait_syncpt( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_wait_syncpt_indx_f(indx) + | host1x_uclass_wait_syncpt_thresh_f(threshold); +} + +static inline u32 host1x_class_host_load_syncpt_base( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_load_syncpt_base_base_indx_f(indx) + | host1x_uclass_load_syncpt_base_value_f(threshold); +} + +static inline u32 host1x_class_host_wait_syncpt_base( + unsigned indx, unsigned base_indx, unsigned offset) +{ + return host1x_uclass_wait_syncpt_base_indx_f(indx) + | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_wait_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt_base( + unsigned base_indx, unsigned offset) +{ + return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_incr_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt( + unsigned cond, unsigned indx) +{ + return host1x_uclass_incr_syncpt_cond_f(cond) + | host1x_uclass_incr_syncpt_indx_f(indx); +} + +static inline u32 host1x_class_host_indoff_reg_write( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indbe_f(0xf) + | host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +static inline u32 host1x_class_host_indoff_reg_read( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset) + | host1x_uclass_indoff_rwn_read_v(); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +static inline u32 host1x_opcode_setclass( + unsigned class_id, unsigned offset, unsigned mask) +{ + return (0 << 28) | (offset << 16) | (class_id << 6) | mask; +} + +static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) +{ + return (1 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) +{ + return (2 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) +{ + return (3 << 28) | (offset << 16) | mask; +} + +static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) +{ + return (4 << 28) | (offset << 16) | value; +} + +static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) +{ + return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), + host1x_class_host_incr_syncpt(cond, indx)); +} + +static inline u32 host1x_opcode_restart(unsigned address) +{ + return (5 << 28) | (address >> 4); +} + +static inline u32 host1x_opcode_gather(unsigned count) +{ + return (6 << 28) | count; +} + +static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | count; +} + +static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; +} + +static inline u32 host1x_opcode_setstreamid(unsigned streamid) +{ + return (7 << 28) | streamid; +} + +static inline u32 host1x_opcode_setpayload(unsigned payload) +{ + return (9 << 28) | payload; +} + +static inline u32 host1x_opcode_gather_wide(unsigned count) +{ + return (12 << 28) | count; +} + +static inline u32 host1x_opcode_acquire_mlock(unsigned mlock) +{ + return (14 << 28) | (0 << 24) | mlock; +} + +static inline u32 host1x_opcode_release_mlock(unsigned mlock) +{ + return (14 << 28) | (1 << 24) | mlock; +} + +#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) + +#endif |