| Message ID | 20190423123122.32573-1-fancer.lancer@gmail.com (mailing list archive) |
|---|---|
| State | Mainlined |
| Commit | 8a08e50cee669387195a1189ec9aee42edc1da49 |
| Headers | show |
| Series | drm: Permit video-buffers writecombine mapping for MIPS | expand |
Am 23.04.19 um 14:31 schrieb Serge Semin: > Since commit 4b050ba7a66c ("MIPS: pgtable.h: Implement the > pgprot_writecombine function for MIPS") and commit c4687b15a848 ("MIPS: Fix > definition of pgprot_writecombine()") write-combine vma mapping is > available to be used by kernel subsystems for MIPS. In particular the > uncached accelerated attribute is requested to be set by ioremap_wc() > method and by generic PCI memory pages/ranges mapping methods. The same > is done by the drm_io_prot()/ttm_io_prot() functions in case if > write-combine flag is set for vma's passed for mapping. But for some > reason the pgprot_writecombine() method calling is ifdefed to be a > platform-specific with MIPS system being marked as lacking of one. At the > very least it doesn't reflect the current MIPS platform implementation. > So in order to improve the DRM subsystem performance on MIPS with UCA > mapping enabled, we need to have pgprot_writecombine() called for buffers, > which need store operations being combined. In case if particular MIPS > chip doesn't support the UCA attribute, the mapping will fall back to > noncached. > > Cc: Ralf Baechle <ralf@linux-mips.org> > Cc: Paul Burton <paul.burton@mips.com> > Cc: James Hogan <jhogan@kernel.org> > Signed-off-by: Vadim V. Vlasov <vadim.vlasov@t-platforms.ru> > Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Not sure about all the details, but at least from the two mile high view that seems to make sense. Acked-by: Christian König <christian.koenig@amd.com> > --- > drivers/gpu/drm/drm_vm.c | 5 +++-- > drivers/gpu/drm/ttm/ttm_bo_util.c | 4 ++-- > 2 files changed, 5 insertions(+), 4 deletions(-) > > diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c > index c3301046dfaa..50178dc64060 100644 > --- a/drivers/gpu/drm/drm_vm.c > +++ b/drivers/gpu/drm/drm_vm.c > @@ -62,7 +62,8 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > /* We don't want graphics memory to be mapped encrypted */ > tmp = pgprot_decrypted(tmp); > > -#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) > +#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) || \ > + defined(__mips__) > if (map->type == _DRM_REGISTERS && !(map->flags & _DRM_WRITE_COMBINING)) > tmp = pgprot_noncached(tmp); > else > @@ -73,7 +74,7 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > tmp = pgprot_writecombine(tmp); > else > tmp = pgprot_noncached(tmp); > -#elif defined(__sparc__) || defined(__arm__) || defined(__mips__) > +#elif defined(__sparc__) || defined(__arm__) > tmp = pgprot_noncached(tmp); > #endif > return tmp; > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c > index 895d77d799e4..9f918b992f7e 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > @@ -539,13 +539,13 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) > tmp = pgprot_noncached(tmp); > #endif > #if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \ > - defined(__powerpc__) > + defined(__powerpc__) || defined(__mips__) > if (caching_flags & TTM_PL_FLAG_WC) > tmp = pgprot_writecombine(tmp); > else > tmp = pgprot_noncached(tmp); > #endif > -#if defined(__sparc__) || defined(__mips__) > +#if defined(__sparc__) > tmp = pgprot_noncached(tmp); > #endif > return tmp;
Hi Serge, On Tue, Apr 23, 2019 at 03:31:22PM +0300, Serge Semin wrote: > Since commit 4b050ba7a66c ("MIPS: pgtable.h: Implement the > pgprot_writecombine function for MIPS") and commit c4687b15a848 ("MIPS: Fix > definition of pgprot_writecombine()") write-combine vma mapping is > available to be used by kernel subsystems for MIPS. In particular the > uncached accelerated attribute is requested to be set by ioremap_wc() > method and by generic PCI memory pages/ranges mapping methods. The same > is done by the drm_io_prot()/ttm_io_prot() functions in case if > write-combine flag is set for vma's passed for mapping. But for some > reason the pgprot_writecombine() method calling is ifdefed to be a > platform-specific with MIPS system being marked as lacking of one. At the > very least it doesn't reflect the current MIPS platform implementation. > So in order to improve the DRM subsystem performance on MIPS with UCA > mapping enabled, we need to have pgprot_writecombine() called for buffers, > which need store operations being combined. In case if particular MIPS > chip doesn't support the UCA attribute, the mapping will fall back to > noncached. > > Cc: Ralf Baechle <ralf@linux-mips.org> > Cc: Paul Burton <paul.burton@mips.com> > Cc: James Hogan <jhogan@kernel.org> > Signed-off-by: Vadim V. Vlasov <vadim.vlasov@t-platforms.ru> > Signed-off-by: Serge Semin <fancer.lancer@gmail.com> > --- > drivers/gpu/drm/drm_vm.c | 5 +++-- > drivers/gpu/drm/ttm/ttm_bo_util.c | 4 ++-- > 2 files changed, 5 insertions(+), 4 deletions(-) Looks good to me: Reviewed-by: Paul Burton <paul.burton@mips.com> Thanks, Paul
Hello folks, Any updates of this patch status? It has been here for about two months. Regards, -Sergey On Tue, Apr 23, 2019 at 03:31:22PM +0300, Serge Semin wrote: > Since commit 4b050ba7a66c ("MIPS: pgtable.h: Implement the > pgprot_writecombine function for MIPS") and commit c4687b15a848 ("MIPS: Fix > definition of pgprot_writecombine()") write-combine vma mapping is > available to be used by kernel subsystems for MIPS. In particular the > uncached accelerated attribute is requested to be set by ioremap_wc() > method and by generic PCI memory pages/ranges mapping methods. The same > is done by the drm_io_prot()/ttm_io_prot() functions in case if > write-combine flag is set for vma's passed for mapping. But for some > reason the pgprot_writecombine() method calling is ifdefed to be a > platform-specific with MIPS system being marked as lacking of one. At the > very least it doesn't reflect the current MIPS platform implementation. > So in order to improve the DRM subsystem performance on MIPS with UCA > mapping enabled, we need to have pgprot_writecombine() called for buffers, > which need store operations being combined. In case if particular MIPS > chip doesn't support the UCA attribute, the mapping will fall back to > noncached. > > Cc: Ralf Baechle <ralf@linux-mips.org> > Cc: Paul Burton <paul.burton@mips.com> > Cc: James Hogan <jhogan@kernel.org> > Signed-off-by: Vadim V. Vlasov <vadim.vlasov@t-platforms.ru> > Signed-off-by: Serge Semin <fancer.lancer@gmail.com> > --- > drivers/gpu/drm/drm_vm.c | 5 +++-- > drivers/gpu/drm/ttm/ttm_bo_util.c | 4 ++-- > 2 files changed, 5 insertions(+), 4 deletions(-) > > diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c > index c3301046dfaa..50178dc64060 100644 > --- a/drivers/gpu/drm/drm_vm.c > +++ b/drivers/gpu/drm/drm_vm.c > @@ -62,7 +62,8 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > /* We don't want graphics memory to be mapped encrypted */ > tmp = pgprot_decrypted(tmp); > > -#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) > +#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) || \ > + defined(__mips__) > if (map->type == _DRM_REGISTERS && !(map->flags & _DRM_WRITE_COMBINING)) > tmp = pgprot_noncached(tmp); > else > @@ -73,7 +74,7 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > tmp = pgprot_writecombine(tmp); > else > tmp = pgprot_noncached(tmp); > -#elif defined(__sparc__) || defined(__arm__) || defined(__mips__) > +#elif defined(__sparc__) || defined(__arm__) > tmp = pgprot_noncached(tmp); > #endif > return tmp; > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c > index 895d77d799e4..9f918b992f7e 100644 > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > @@ -539,13 +539,13 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) > tmp = pgprot_noncached(tmp); > #endif > #if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \ > - defined(__powerpc__) > + defined(__powerpc__) || defined(__mips__) > if (caching_flags & TTM_PL_FLAG_WC) > tmp = pgprot_writecombine(tmp); > else > tmp = pgprot_noncached(tmp); > #endif > -#if defined(__sparc__) || defined(__mips__) > +#if defined(__sparc__) > tmp = pgprot_noncached(tmp); > #endif > return tmp; > -- > 2.21.0 >
On Mon, Jun 17, 2019 at 04:47:30PM +0300, Serge Semin wrote: > Hello folks, > > Any updates of this patch status? It has been here for about two months. > Sorry for the mixup, looks like this one just fell through the cracks. I've applied it to drm-misc-next with the attached Ack and Review. Sean > Regards, > -Sergey > > On Tue, Apr 23, 2019 at 03:31:22PM +0300, Serge Semin wrote: > > Since commit 4b050ba7a66c ("MIPS: pgtable.h: Implement the > > pgprot_writecombine function for MIPS") and commit c4687b15a848 ("MIPS: Fix > > definition of pgprot_writecombine()") write-combine vma mapping is > > available to be used by kernel subsystems for MIPS. In particular the > > uncached accelerated attribute is requested to be set by ioremap_wc() > > method and by generic PCI memory pages/ranges mapping methods. The same > > is done by the drm_io_prot()/ttm_io_prot() functions in case if > > write-combine flag is set for vma's passed for mapping. But for some > > reason the pgprot_writecombine() method calling is ifdefed to be a > > platform-specific with MIPS system being marked as lacking of one. At the > > very least it doesn't reflect the current MIPS platform implementation. > > So in order to improve the DRM subsystem performance on MIPS with UCA > > mapping enabled, we need to have pgprot_writecombine() called for buffers, > > which need store operations being combined. In case if particular MIPS > > chip doesn't support the UCA attribute, the mapping will fall back to > > noncached. > > > > Cc: Ralf Baechle <ralf@linux-mips.org> > > Cc: Paul Burton <paul.burton@mips.com> > > Cc: James Hogan <jhogan@kernel.org> > > Signed-off-by: Vadim V. Vlasov <vadim.vlasov@t-platforms.ru> > > Signed-off-by: Serge Semin <fancer.lancer@gmail.com> > > --- > > drivers/gpu/drm/drm_vm.c | 5 +++-- > > drivers/gpu/drm/ttm/ttm_bo_util.c | 4 ++-- > > 2 files changed, 5 insertions(+), 4 deletions(-) > > > > diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c > > index c3301046dfaa..50178dc64060 100644 > > --- a/drivers/gpu/drm/drm_vm.c > > +++ b/drivers/gpu/drm/drm_vm.c > > @@ -62,7 +62,8 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > > /* We don't want graphics memory to be mapped encrypted */ > > tmp = pgprot_decrypted(tmp); > > > > -#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) > > +#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) || \ > > + defined(__mips__) > > if (map->type == _DRM_REGISTERS && !(map->flags & _DRM_WRITE_COMBINING)) > > tmp = pgprot_noncached(tmp); > > else > > @@ -73,7 +74,7 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > > tmp = pgprot_writecombine(tmp); > > else > > tmp = pgprot_noncached(tmp); > > -#elif defined(__sparc__) || defined(__arm__) || defined(__mips__) > > +#elif defined(__sparc__) || defined(__arm__) > > tmp = pgprot_noncached(tmp); > > #endif > > return tmp; > > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c > > index 895d77d799e4..9f918b992f7e 100644 > > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > > @@ -539,13 +539,13 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) > > tmp = pgprot_noncached(tmp); > > #endif > > #if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \ > > - defined(__powerpc__) > > + defined(__powerpc__) || defined(__mips__) > > if (caching_flags & TTM_PL_FLAG_WC) > > tmp = pgprot_writecombine(tmp); > > else > > tmp = pgprot_noncached(tmp); > > #endif > > -#if defined(__sparc__) || defined(__mips__) > > +#if defined(__sparc__) > > tmp = pgprot_noncached(tmp); > > #endif > > return tmp; > > -- > > 2.21.0 > >
Hello Sean, On Tue, Jun 18, 2019 at 03:52:04PM -0400, Sean Paul wrote: > On Mon, Jun 17, 2019 at 04:47:30PM +0300, Serge Semin wrote: > > Hello folks, > > > > Any updates of this patch status? It has been here for about two months. > > > > Sorry for the mixup, looks like this one just fell through the cracks. I've > applied it to drm-misc-next with the attached Ack and Review. > Great! Thank you. Regards, -Sergey > Sean > > > > Regards, > > -Sergey > > > > On Tue, Apr 23, 2019 at 03:31:22PM +0300, Serge Semin wrote: > > > Since commit 4b050ba7a66c ("MIPS: pgtable.h: Implement the > > > pgprot_writecombine function for MIPS") and commit c4687b15a848 ("MIPS: Fix > > > definition of pgprot_writecombine()") write-combine vma mapping is > > > available to be used by kernel subsystems for MIPS. In particular the > > > uncached accelerated attribute is requested to be set by ioremap_wc() > > > method and by generic PCI memory pages/ranges mapping methods. The same > > > is done by the drm_io_prot()/ttm_io_prot() functions in case if > > > write-combine flag is set for vma's passed for mapping. But for some > > > reason the pgprot_writecombine() method calling is ifdefed to be a > > > platform-specific with MIPS system being marked as lacking of one. At the > > > very least it doesn't reflect the current MIPS platform implementation. > > > So in order to improve the DRM subsystem performance on MIPS with UCA > > > mapping enabled, we need to have pgprot_writecombine() called for buffers, > > > which need store operations being combined. In case if particular MIPS > > > chip doesn't support the UCA attribute, the mapping will fall back to > > > noncached. > > > > > > Cc: Ralf Baechle <ralf@linux-mips.org> > > > Cc: Paul Burton <paul.burton@mips.com> > > > Cc: James Hogan <jhogan@kernel.org> > > > Signed-off-by: Vadim V. Vlasov <vadim.vlasov@t-platforms.ru> > > > Signed-off-by: Serge Semin <fancer.lancer@gmail.com> > > > --- > > > drivers/gpu/drm/drm_vm.c | 5 +++-- > > > drivers/gpu/drm/ttm/ttm_bo_util.c | 4 ++-- > > > 2 files changed, 5 insertions(+), 4 deletions(-) > > > > > > diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c > > > index c3301046dfaa..50178dc64060 100644 > > > --- a/drivers/gpu/drm/drm_vm.c > > > +++ b/drivers/gpu/drm/drm_vm.c > > > @@ -62,7 +62,8 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > > > /* We don't want graphics memory to be mapped encrypted */ > > > tmp = pgprot_decrypted(tmp); > > > > > > -#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) > > > +#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) || \ > > > + defined(__mips__) > > > if (map->type == _DRM_REGISTERS && !(map->flags & _DRM_WRITE_COMBINING)) > > > tmp = pgprot_noncached(tmp); > > > else > > > @@ -73,7 +74,7 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, > > > tmp = pgprot_writecombine(tmp); > > > else > > > tmp = pgprot_noncached(tmp); > > > -#elif defined(__sparc__) || defined(__arm__) || defined(__mips__) > > > +#elif defined(__sparc__) || defined(__arm__) > > > tmp = pgprot_noncached(tmp); > > > #endif > > > return tmp; > > > diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c > > > index 895d77d799e4..9f918b992f7e 100644 > > > --- a/drivers/gpu/drm/ttm/ttm_bo_util.c > > > +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c > > > @@ -539,13 +539,13 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) > > > tmp = pgprot_noncached(tmp); > > > #endif > > > #if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \ > > > - defined(__powerpc__) > > > + defined(__powerpc__) || defined(__mips__) > > > if (caching_flags & TTM_PL_FLAG_WC) > > > tmp = pgprot_writecombine(tmp); > > > else > > > tmp = pgprot_noncached(tmp); > > > #endif > > > -#if defined(__sparc__) || defined(__mips__) > > > +#if defined(__sparc__) > > > tmp = pgprot_noncached(tmp); > > > #endif > > > return tmp; > > > -- > > > 2.21.0 > > > > > -- > Sean Paul, Software Engineer, Google / Chromium OS
diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c index c3301046dfaa..50178dc64060 100644 --- a/drivers/gpu/drm/drm_vm.c +++ b/drivers/gpu/drm/drm_vm.c @@ -62,7 +62,8 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, /* We don't want graphics memory to be mapped encrypted */ tmp = pgprot_decrypted(tmp); -#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) +#if defined(__i386__) || defined(__x86_64__) || defined(__powerpc__) || \ + defined(__mips__) if (map->type == _DRM_REGISTERS && !(map->flags & _DRM_WRITE_COMBINING)) tmp = pgprot_noncached(tmp); else @@ -73,7 +74,7 @@ static pgprot_t drm_io_prot(struct drm_local_map *map, tmp = pgprot_writecombine(tmp); else tmp = pgprot_noncached(tmp); -#elif defined(__sparc__) || defined(__arm__) || defined(__mips__) +#elif defined(__sparc__) || defined(__arm__) tmp = pgprot_noncached(tmp); #endif return tmp; diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index 895d77d799e4..9f918b992f7e 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -539,13 +539,13 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) tmp = pgprot_noncached(tmp); #endif #if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \ - defined(__powerpc__) + defined(__powerpc__) || defined(__mips__) if (caching_flags & TTM_PL_FLAG_WC) tmp = pgprot_writecombine(tmp); else tmp = pgprot_noncached(tmp); #endif -#if defined(__sparc__) || defined(__mips__) +#if defined(__sparc__) tmp = pgprot_noncached(tmp); #endif return tmp;
Since commit 4b050ba7a66c ("MIPS: pgtable.h: Implement the pgprot_writecombine function for MIPS") and commit c4687b15a848 ("MIPS: Fix definition of pgprot_writecombine()") write-combine vma mapping is available to be used by kernel subsystems for MIPS. In particular the uncached accelerated attribute is requested to be set by ioremap_wc() method and by generic PCI memory pages/ranges mapping methods. The same is done by the drm_io_prot()/ttm_io_prot() functions in case if write-combine flag is set for vma's passed for mapping. But for some reason the pgprot_writecombine() method calling is ifdefed to be a platform-specific with MIPS system being marked as lacking of one. At the very least it doesn't reflect the current MIPS platform implementation. So in order to improve the DRM subsystem performance on MIPS with UCA mapping enabled, we need to have pgprot_writecombine() called for buffers, which need store operations being combined. In case if particular MIPS chip doesn't support the UCA attribute, the mapping will fall back to noncached. Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Paul Burton <paul.burton@mips.com> Cc: James Hogan <jhogan@kernel.org> Signed-off-by: Vadim V. Vlasov <vadim.vlasov@t-platforms.ru> Signed-off-by: Serge Semin <fancer.lancer@gmail.com> --- drivers/gpu/drm/drm_vm.c | 5 +++-- drivers/gpu/drm/ttm/ttm_bo_util.c | 4 ++-- 2 files changed, 5 insertions(+), 4 deletions(-)