| 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: <000a01ce7e2d$9de742d0$d9b5c870$@samsung.com>
Date: Thu, 11 Jul 2013 20:56:02 +0900
From: Cho KyongHo <pullip.cho@...sung.com>
To: 'Prathyush K' <prathyush@...omium.org>
Cc: 'Linux ARM Kernel' <linux-arm-kernel@...ts.infradead.org>,
'Linux IOMMU' <iommu@...ts.linux-foundation.org>,
'Linux Kernel' <linux-kernel@...r.kernel.org>,
'Linux Samsung SOC' <linux-samsung-soc@...r.kernel.org>,
'Hyunwoong Kim' <khw0178.kim@...sung.com>,
'Joerg Roedel' <joro@...tes.org>,
'Kukjin Kim' <kgene.kim@...sung.com>,
'Prathyush' <prathyush.k@...sung.com>,
'Rahul Sharma' <rahul.sharma@...sung.com>,
'Subash Patel' <supash.ramaswamy@...aro.org>,
'Keyyoung Park' <keyyoung.park@...sung.com>,
'Grant Grundler' <grundler@...omium.org>,
'Thomas Abraham' <thomas.abraham@...aro.org>
Subject: RE: [PATCH v7 6/9] clk: exynos5250: add gate clock descriptions of
System MMU
> From: Prathyush K [mailto:prathyush@...omium.org]
> Sent: Wednesday, July 10, 2013 11:49 PM
>
> On Fri, Jul 5, 2013 at 5:59 PM, Cho KyongHo <pullip.cho@...sung.com> wrote:
> This adds gate clocks of all System MMUs and their master IPs
> that are not apeared in clk-exynos5250.c
>
> Signed-off-by: Cho KyongHo <pullip.cho@...sung.com>
> CC: Thomas Abraham <thomas.abraham@...aro.org>
> ---
> .../devicetree/bindings/clock/exynos5250-clock.txt | 28 +++++++++-
> drivers/clk/samsung/clk-exynos5250.c | 57 ++++++++++++++++---
> 2 files changed, 75 insertions(+), 10 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/clock/exynos5250-clock.txt
> b/Documentation/devicetree/bindings/clock/exynos5250-clock.txt
> index 781a627..df49694 100644
> --- a/Documentation/devicetree/bindings/clock/exynos5250-clock.txt
> +++ b/Documentation/devicetree/bindings/clock/exynos5250-clock.txt
> @@ -154,7 +154,33 @@ clock which they consume.
> dsim0 341
> dp 342
> mixer 343
> - hdmi 345
> + hdmi 344
> + camif_top 345
> + smmu_fimc_lite0 346
> + smmu_fimc_lite1 347
> + smmu_fimc_lite2 348
> + smmu_tv 349
> + smmu_fimd1 350
> + smmu_2d 351
> + fimc_isp 352
> + fimc_drc 353
> + fimc_fd 354
> + fimc_scc 355
> + fimc_scp 356
> + fimc_mcuctl 357
> + fimc_odc 358
> + fimc_dis 359
> + fimc_3dnr 360
> + smmu_fimc_isp 361
> + smmu_fimc_drc 362
> + smmu_fimc_fd 363
> + smmu_fimc_scc 364
> + smmu_fimc_scp 365
> + smmu_fimc_mcuctl 366
> + smmu_fimc_odc 367
> + smmu_fimc_dis0 368
> + smmu_fimc_dis1 369
> + smmu_fimc_3dnr 370
>
> Example 1: An example of a clock controller node is listed below.
>
> diff --git a/drivers/clk/samsung/clk-exynos5250.c b/drivers/clk/samsung/clk-exynos5250.c
> index 22d7699..1921d6c 100644
> --- a/drivers/clk/samsung/clk-exynos5250.c
> +++ b/drivers/clk/samsung/clk-exynos5250.c
> @@ -53,12 +53,15 @@
> #define DIV_PERIC3 0x10564
> #define DIV_PERIC4 0x10568
> #define DIV_PERIC5 0x1056c
> +#define GATE_IP_ISP0 0x0C800
> +#define GATE_IP_ISP1 0x0C800
> #define GATE_IP_GSCL 0x10920
> #define GATE_IP_MFC 0x1092c
> #define GATE_IP_GEN 0x10934
> #define GATE_IP_FSYS 0x10944
> #define GATE_IP_PERIC 0x10950
> #define GATE_IP_PERIS 0x10960
> +#define GATE_IP_ACP 0x18800
> #define SRC_CDREX 0x20200
> #define PLL_DIV2_SEL 0x20a24
> #define GATE_IP_DISP1 0x10928
> @@ -100,6 +103,14 @@ enum exynos5250_clks {
> tzpc2, tzpc3, tzpc4, tzpc5, tzpc6, tzpc7, tzpc8, tzpc9, hdmi_cec, mct,
> wdt, rtc, tmu, fimd1, mie1, dsim0, dp, mixer, hdmi,
>
> + camif_top, smmu_fimc_lite0, smmu_fimc_lite1, smmu_fimc_lite2,
> + smmu_tv, smmu_fimd1, smmu_2d,
> + fimc_isp, fimc_drc, fimc_fd, fimc_scc, fimc_scp, fimc_mcuctl, fimc_odc,
> + fimc_dis, fimc_3dnr,
> + smmu_fimc_isp, smmu_fimc_drc, smmu_fimc_fd, smmu_fimc_scc,
> + smmu_fimc_scp, smmu_fimc_mcuctl, smmu_fimc_odc, smmu_fimc_dis0,
> + smmu_fimc_dis1, smmu_fimc_3dnr,
> +
> nr_clks,
> };
>
> @@ -320,19 +331,26 @@ struct samsung_gate_clock exynos5250_gate_clks[] __initdata = {
> GATE(gscl3, "gscl3", "aclk266", GATE_IP_GSCL, 3, 0, 0),
> GATE(gscl_wa, "gscl_wa", "div_gscl_wa", GATE_IP_GSCL, 5, 0, 0),
> GATE(gscl_wb, "gscl_wb", "div_gscl_wb", GATE_IP_GSCL, 6, 0, 0),
> - GATE(smmu_gscl0, "smmu_gscl0", "aclk266", GATE_IP_GSCL, 7, 0, 0),
> - GATE(smmu_gscl1, "smmu_gscl1", "aclk266", GATE_IP_GSCL, 8, 0, 0),
> - GATE(smmu_gscl2, "smmu_gscl2", "aclk266", GATE_IP_GSCL, 9, 0, 0),
> - GATE(smmu_gscl3, "smmu_gscl3", "aclk266", GATE_IP_GSCL, 10, 0, 0),
> + GATE(smmu_gscl0, "smmu_gscl0", "gscl0", GATE_IP_GSCL, 7, 0, 0),
> + GATE(smmu_gscl1, "smmu_gscl1", "gscl1", GATE_IP_GSCL, 8, 0, 0),
> + GATE(smmu_gscl2, "smmu_gscl2", "gscl2", GATE_IP_GSCL, 9, 0, 0),
> + GATE(smmu_gscl3, "smmu_gscl3", "gscl3", GATE_IP_GSCL, 10, 0, 0),
> + GATE(camif_top, "camif_top", "aclk266", GATE_IP_GSCL, 4, 0, 0),
> + GATE(smmu_fimc_lite0, "smmu_fimc_lite0", "camif_top",
> + GATE_IP_GSCL, 12, 0, 0),
> + GATE(smmu_fimc_lite1, "smmu_fimc_lite1", "camif_top",
> + GATE_IP_GSCL, 13, 0, 0),
> + GATE(smmu_fimc_lite2, "smmu_fimc_lite2", "camif_top",
> + GATE_IP_GSCL, 14, 0, 0),
> GATE(mfc, "mfc", "aclk333", GATE_IP_MFC, 0, 0, 0),
> - GATE(smmu_mfcl, "smmu_mfcl", "aclk333", GATE_IP_MFC, 1, 0, 0),
> - GATE(smmu_mfcr, "smmu_mfcr", "aclk333", GATE_IP_MFC, 2, 0, 0),
> + GATE(smmu_mfcl, "smmu_mfcl", "mfc", GATE_IP_MFC, 1, 0, 0),
> + GATE(smmu_mfcr, "smmu_mfcr", "mfc", GATE_IP_MFC, 2, 0, 0),
>
> This should be reversed as seen in 5250. The mfcl's control bit is 2 and mfcr's control bit is 1.
>
Oh. Really it is.
I didn't check it.
I remember that you have pointed that several times already. :)
> Why is the parent of the smmu_mfc set as "mfc"? Is this a required hierarchy in the clocks?
> Even I noticed that we need to turn on MFC's clock to be able to access smmu_mfc's registers.
> But I don't think setting mfc as parent of smmu_mfc is the correct solution since this results in MFC not getting clock gated
during video playback.
>
> Beginning of playback:
> Turn on SYSMMU:
> MFC Sysmmu clock: +1
> MFC clock: +1
>
> During video playback, dynamic clock gating of MFC:
> Turn on MFC (before frame decode)
> MFC Sysmmu clock: +1
> MFC clock: +2
>
> Turn off MFC (after frame decode)
> MFC Sysmmu clock: +1
> MFC clock: +1
>
> End of playback:
> Turn off SYSMMU:
> MFC Sysmmu clock: 0 (gated)
> MFC clock: 0 (gated)
>
> As can be seen, MFC's dynamic clock gating between frames never works since the sysmmu has already turned on the clock.
>
> A better way is to pass the mmu-master's clock through the device node (optional parameter).
> If the parent device's clock is present in the sysmmu node, we turn ON/OFF this clock before/after we access any sysmmu registers
> E.g. before/after TLB invalidation.
>
> What do you think?
>
Hmm,, it is also a good idea to handle master's clock explicitly
Master's clock is already passed to the System MMU driver if it is required
via dts.
Please let me think about it again
Thank you.
Cho KyongHo.
--
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