Re: [PATCH 6/6] vduse: Support specifying bounce buffer size via sysfs

2022-11-16 Thread Jason Wang
On Thu, Nov 17, 2022 at 2:08 PM Yongji Xie wrote: > > On Thu, Nov 17, 2022 at 1:26 PM Jason Wang wrote: > > > > On Mon, Nov 14, 2022 at 3:16 PM Xie Yongji wrote: > > > > > > Add sysfs interface to support specifying bounce > > > buffer size in virtio-vdpa case. This is a performance > > > tuning

Re: [PATCH 3/6] vduse: Add sysfs interface for irq affinity setup

2022-11-16 Thread Jason Wang
On Thu, Nov 17, 2022 at 1:48 PM Yongji Xie wrote: > > On Thu, Nov 17, 2022 at 11:37 AM Jason Wang wrote: > > > > On Wed, Nov 16, 2022 at 3:46 PM Yongji Xie wrote: > > > > > > On Wed, Nov 16, 2022 at 3:11 PM Jason Wang wrote: > > > > > > > > On Tue, Nov 15, 2022 at 10:49 AM Yongji Xie > > > >

Re: [PATCH] tools/virtio: fix vrinth_test hang

2022-11-16 Thread Jason Wang
在 2022/11/15 19:17, Liming Wu 写道: vringh_test hangs on __vring_new_virtqueue() because `vqs_list_lock` is not initialized. Let's initialize it before call __vring_new_virtqueue() Signed-off-by: Liming Wu Acked-by: Jason Wang Thanks --- tools/virtio/vringh_test.c | 2 ++ 1 file cha

Re: [PATCH 6/6] vduse: Support specifying bounce buffer size via sysfs

2022-11-16 Thread Jason Wang
On Mon, Nov 14, 2022 at 3:16 PM Xie Yongji wrote: > > Add sysfs interface to support specifying bounce > buffer size in virtio-vdpa case. This is a performance > tuning parameter for high throughput workloads. I wonder what's the reason for not having this in VDUSE_CREATE_DEV? Thanks > > Signed

Re: [PATCH 1/6] vduse: Refactor allocation for vduse virtqueues

2022-11-16 Thread Jason Wang
On Mon, Nov 14, 2022 at 3:16 PM Xie Yongji wrote: > > Allocate memory for vduse virtqueues one by one instead of > doing one allocation for all of them. > > This is a preparation for adding sysfs interface for virtqueues. The code looks but I think it's not a must since each kobject could be allo

Re: [PATCH 3/6] vduse: Add sysfs interface for irq affinity setup

2022-11-16 Thread Jason Wang
On Wed, Nov 16, 2022 at 3:46 PM Yongji Xie wrote: > > On Wed, Nov 16, 2022 at 3:11 PM Jason Wang wrote: > > > > On Tue, Nov 15, 2022 at 10:49 AM Yongji Xie wrote: > > > > > > On Mon, Nov 14, 2022 at 4:55 PM Jason Wang wrote: > > > > > > > > On Mon, Nov 14, 2022 at 4:20 PM Yongji Xie > > > > w

[PATCH V2] vdpa: allow provisioning device features

2022-11-16 Thread Jason Wang
This patch allows device features to be provisioned via vdpa. This will be useful for preserving migration compatibility between source and destination: # vdpa dev add name dev1 mgmtdev pci/:02:00.0 device_features 0x30002 # dev1: mac 52:54:00:12:34:56 link up link_announce false mtu 65535

[PATCH 6/7] drm/solomon: Set preferred color depth and bpp to the correct values

2022-11-16 Thread Thomas Zimmermann
Set the preferred color depth to 24 bits and the fbdev bpp to 32 bits. This will signal XRGB as default format to clients. Signed-off-by: Thomas Zimmermann --- drivers/gpu/drm/solomon/ssd130x.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/solomon/ss

[PATCH 5/7] drm/simpledrm: Set preferred depth from format of scanout buffer

2022-11-16 Thread Thomas Zimmermann
Set the preferred depth from the format of the scanout buffer. The value cannot be hardcoded, as the scanout buffer's is only known at runtime. Also derive the fbdev emulations bpp value from the scanout format. Signed-off-by: Thomas Zimmermann --- drivers/gpu/drm/tiny/simpledrm.c | 4 ++-- 1 fi

[PATCH 3/7] drm/cirrus: Decouple fbdev bpp value from color depth

2022-11-16 Thread Thomas Zimmermann
Cirrus has a preferred color depth of 16 bit; also use it as fbdev bpp value. Don't use the color depth directly. It has a different meaning than bpp and both cannot be used interchangeably. Signed-off-by: Thomas Zimmermann --- drivers/gpu/drm/tiny/cirrus.c | 2 +- 1 file changed, 1 insertion(+)

[PATCH 7/7] drm/fb-helper: Don't use the preferred depth for the BPP default

2022-11-16 Thread Thomas Zimmermann
If no preferred value for bits-per-pixel has been given, fall back to 32. Never use the preferred depth. The color depth is the number of color/alpha bits per pixel, while bpp is the overall number of bits in most cases. Most noteworthy, XRGB has a depth of 24 and a bpp value of 32. Using dept

[PATCH 2/7] drm/logicvc: Fix preferred fbdev cpp

2022-11-16 Thread Thomas Zimmermann
Logicvc can have different values for the preferred color depth. Set the fbdev bpp value depending on the runtime value. Signed-off-by: Thomas Zimmermann --- drivers/gpu/drm/logicvc/logicvc_drm.c | 14 +- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/l

[PATCH 4/7] drm/ofdrm: Set preferred depth from format of scanout buffer

2022-11-16 Thread Thomas Zimmermann
Set the preferred depth from the format of the scanout buffer. The value cannot be hardcoded, as the scanout buffer's is only known at runtime. Keeping the existing switch statement just duplicates the driver's existing logic for format detection. Also remove the FIXME comment from the call to drm

[PATCH 0/7] drm: Fix the color-depth/bpp confusion

2022-11-16 Thread Thomas Zimmermann
A number of drivers mix up the meaning of bits-per-pixel and color depth. For each of them, set the correct values. As a rule of thumb, the color depth is the number of color and alpha bits that affect image composition. The bpp value is the color depth in the pixel plus the filler bits. The color

[PATCH 1/7] drm/hisilicon/hibmc: Fix preferred depth and bpp

2022-11-16 Thread Thomas Zimmermann
Set the preferred color depth to 24 bits and the fbdev bpp to 32 bits. This will signal XRGB as default format to clients. Signed-off-by: Thomas Zimmermann --- drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/

Re: [PATCH v2 12/44] cpuidle,dt: Push RCU-idle into driver

2022-11-16 Thread Peter Zijlstra
Sorry; things keep getting in the way of finishing this :/ As such, I need a bit of time to get on-track again.. On Tue, Oct 04, 2022 at 01:03:57PM +0200, Ulf Hansson wrote: > > --- a/drivers/acpi/processor_idle.c > > +++ b/drivers/acpi/processor_idle.c > > @@ -1200,6 +1200,8 @@ static int acp

Re: [PATCH v2] x86/paravirt: use common macro for creating simple asm paravirt functions

2022-11-16 Thread Juergen Gross via Virtualization
On 16.11.22 12:04, Peter Zijlstra wrote: On Wed, Nov 09, 2022 at 02:44:18PM +0100, Juergen Gross wrote: There are some paravirt assembler functions which are sharing a common pattern. Introduce a macro DEFINE_PARAVIRT_ASM() for creating them. Note that this macro is including explicit alignment

Re: [PATCH v2] x86/paravirt: use common macro for creating simple asm paravirt functions

2022-11-16 Thread Peter Zijlstra
On Wed, Nov 09, 2022 at 02:44:18PM +0100, Juergen Gross wrote: > There are some paravirt assembler functions which are sharing a common > pattern. Introduce a macro DEFINE_PARAVIRT_ASM() for creating them. > > Note that this macro is including explicit alignment of the generated > functions, leadi