| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CADnq5_NiY0MZz0ohp7aoOGu_tZLNYfGqabVAoXNUHMX9wcraGg@mail.gmail.com>
Date: Tue, 14 Oct 2014 12:50:28 -0400
From: Alex Deucher <alexdeucher@...il.com>
To: Oded Gabbay <oded.gabbay@....com>
Cc: David Airlie <airlied@...ux.ie>,
Alex Deucher <alexander.deucher@....com>,
Jerome Glisse <j.glisse@...il.com>,
Andrew Lewycky <Andrew.Lewycky@....com>,
LKML <linux-kernel@...r.kernel.org>,
Maling list - DRI developers
<dri-devel@...ts.freedesktop.org>,
Christian König <christian.koenig@....com>
Subject: Re: [PATCH v4 01/23] drm/radeon: reduce number of free VMIDs and
pipes in KV
On Wed, Sep 24, 2014 at 4:45 PM, Oded Gabbay <oded.gabbay@....com> wrote:
> To support HSA on KV, we need to limit the number of vmids and pipes
> that are available for radeon's use with KV.
>
> This patch reserves VMIDs 8-15 for amdkfd (so radeon can only use VMIDs
> 0-7) and also makes radeon thinks that KV has only a single MEC with a single
> pipe in it
>
> v3: Use define for static vmid allocation in radeon
>
> Signed-off-by: Oded Gabbay <oded.gabbay@....com>
Reviewed-by: Alex Deucher <alexander.deucher@....com>
> ---
> drivers/gpu/drm/radeon/cik.c | 48 +++++++++++++++++++++----------------------
> drivers/gpu/drm/radeon/cikd.h | 2 ++
> 2 files changed, 26 insertions(+), 24 deletions(-)
>
> diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
> index 0d761f7..1530e37 100644
> --- a/drivers/gpu/drm/radeon/cik.c
> +++ b/drivers/gpu/drm/radeon/cik.c
> @@ -4678,12 +4678,11 @@ static int cik_mec_init(struct radeon_device *rdev)
> /*
> * KV: 2 MEC, 4 Pipes/MEC, 8 Queues/Pipe - 64 Queues total
> * CI/KB: 1 MEC, 4 Pipes/MEC, 8 Queues/Pipe - 32 Queues total
> + * Nonetheless, we assign only 1 pipe because all other pipes will
> + * be handled by KFD
> */
> - if (rdev->family == CHIP_KAVERI)
> - rdev->mec.num_mec = 2;
> - else
> - rdev->mec.num_mec = 1;
> - rdev->mec.num_pipe = 4;
> + rdev->mec.num_mec = 1;
> + rdev->mec.num_pipe = 1;
> rdev->mec.num_queue = rdev->mec.num_mec * rdev->mec.num_pipe * 8;
>
> if (rdev->mec.hpd_eop_obj == NULL) {
> @@ -4825,28 +4824,24 @@ static int cik_cp_compute_resume(struct radeon_device *rdev)
>
> /* init the pipes */
> mutex_lock(&rdev->srbm_mutex);
> - for (i = 0; i < (rdev->mec.num_pipe * rdev->mec.num_mec); i++) {
> - int me = (i < 4) ? 1 : 2;
> - int pipe = (i < 4) ? i : (i - 4);
>
> - eop_gpu_addr = rdev->mec.hpd_eop_gpu_addr + (i * MEC_HPD_SIZE * 2);
> + eop_gpu_addr = rdev->mec.hpd_eop_gpu_addr;
>
> - cik_srbm_select(rdev, me, pipe, 0, 0);
> + cik_srbm_select(rdev, 0, 0, 0, 0);
>
> - /* write the EOP addr */
> - WREG32(CP_HPD_EOP_BASE_ADDR, eop_gpu_addr >> 8);
> - WREG32(CP_HPD_EOP_BASE_ADDR_HI, upper_32_bits(eop_gpu_addr) >> 8);
> + /* write the EOP addr */
> + WREG32(CP_HPD_EOP_BASE_ADDR, eop_gpu_addr >> 8);
> + WREG32(CP_HPD_EOP_BASE_ADDR_HI, upper_32_bits(eop_gpu_addr) >> 8);
>
> - /* set the VMID assigned */
> - WREG32(CP_HPD_EOP_VMID, 0);
> + /* set the VMID assigned */
> + WREG32(CP_HPD_EOP_VMID, 0);
> +
> + /* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
> + tmp = RREG32(CP_HPD_EOP_CONTROL);
> + tmp &= ~EOP_SIZE_MASK;
> + tmp |= order_base_2(MEC_HPD_SIZE / 8);
> + WREG32(CP_HPD_EOP_CONTROL, tmp);
>
> - /* set the EOP size, register value is 2^(EOP_SIZE+1) dwords */
> - tmp = RREG32(CP_HPD_EOP_CONTROL);
> - tmp &= ~EOP_SIZE_MASK;
> - tmp |= order_base_2(MEC_HPD_SIZE / 8);
> - WREG32(CP_HPD_EOP_CONTROL, tmp);
> - }
> - cik_srbm_select(rdev, 0, 0, 0, 0);
> mutex_unlock(&rdev->srbm_mutex);
>
> /* init the queues. Just two for now. */
> @@ -5900,8 +5895,13 @@ int cik_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib)
> */
> int cik_vm_init(struct radeon_device *rdev)
> {
> - /* number of VMs */
> - rdev->vm_manager.nvm = 16;
> + /*
> + * number of VMs
> + * VMID 0 is reserved for System
> + * radeon graphics/compute will use VMIDs 1-7
> + * amdkfd will use VMIDs 8-15
> + */
> + rdev->vm_manager.nvm = RADEON_NUM_OF_VMIDS;
> /* base offset of vram pages */
> if (rdev->flags & RADEON_IS_IGP) {
> u64 tmp = RREG32(MC_VM_FB_OFFSET);
> diff --git a/drivers/gpu/drm/radeon/cikd.h b/drivers/gpu/drm/radeon/cikd.h
> index 0c6e1b5..fae4d0c 100644
> --- a/drivers/gpu/drm/radeon/cikd.h
> +++ b/drivers/gpu/drm/radeon/cikd.h
> @@ -30,6 +30,8 @@
> #define CIK_RB_BITMAP_WIDTH_PER_SH 2
> #define HAWAII_RB_BITMAP_WIDTH_PER_SH 4
>
> +#define RADEON_NUM_OF_VMIDS 8
> +
> /* DIDT IND registers */
> #define DIDT_SQ_CTRL0 0x0
> # define DIDT_CTRL_EN (1 << 0)
> --
> 1.9.1
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@...ts.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/dri-devel
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists