Paul, On Sunday 18 September 2011 02:28 PM, Santosh wrote: > On Friday 16 September 2011 09:10 PM, Cousson, Benoit wrote: >> Hi Sricharan, >> >> On 9/9/2011 6:02 PM, R, Sricharan wrote: >>> The address spaces, irqs and dma reqs count API returns the >>> number of corresponding entries in a hwmod including a additional >>> null value or a -1 terminator in the structure introduced >>> recently. More information here: >>> >>> 212738a4 (omap_hwmod: use a terminator record with >>> omap_hwmod_mpu_irqs arrays) >>> 78183f3f (omap_hwmod: use a null structure record to terminate >>> omap_hwmod_addr_space arrays) >>> bc614958 (omap_hwmod: use a terminator record with >>> omap_hwmod_dma_info arrays) >>> >>> The devices which have multiple hwmods and use device_build_ss are >>> broken with this, as their resources are populated with a >>> extra null value, subsequently the probe fails. So fix the API not to >>> include the array terminator in the count. >>> >>> Signed-off-by: sricharan<r.sricharan@xxxxxx> >>> Signed-off-by: Santosh Shilimkar<santosh.shilimkar@xxxxxx> >>> Cc: Benoit Cousson<b-cousson@xxxxxx> >>> Cc: Paul Walmsley<paul@xxxxxxxxx> >>> Cc: Kevin Hilman<khilman@xxxxxx> >> >> It would have been nice to give me the credit for the bug report on the >> other functions, but otherwise: >> > Sure. Will add reported-by of yours. > >> Acked-by: Benoit Cousson <b-cousson@xxxxxx> >> > Thanks > >> And I still prefer the return i - 1 :-) >> > Let's do "i-1" as you prefer. Maintainer > always have final say !! > Here is the updated patch with i-1 as agreed with Benoit. Also attaching it in case mailer eats spaces. >From c55f38512e1471b8a9405c2cb45d9377e0c45999 Mon Sep 17 00:00:00 2001 From: sricharan <r.sricharan@xxxxxx> Date: Wed, 21 Sep 2011 16:08:46 +0530 Subject: [PATCH 1/8] OMAP: hwmod: Fix the addr space, irq, dma count APIs The address spaces, irqs and dma reqs count API returns the number of corresponding entries in a hwmod including a additional null value or a -1 terminator in the structure introduced recently. More information here: - 212738a4 (omap_hwmod: use a terminator record with omap_hwmod_mpu_irqs arrays) - 78183f3f (omap_hwmod: use a null structure record to terminate omap_hwmod_addr_space arrays) - bc614958 (omap_hwmod: use a terminator record with omap_hwmod_dma_info arrays) The issue with irqs and dma info was originally reported by Benoit Cousson. The devices which have multiple hwmods and use device_build_ss are broken with this, as their resources are populated with a extra null value, subsequently the probe fails. So fix the API not to include the array terminator in the count. Reported-by: Benoit Cousson <b-cousson@xxxxxx> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@xxxxxx> Signed-off-by: sricharan <r.sricharan@xxxxxx> Acked: Benoit Cousson <b-cousson@xxxxxx> --- arch/arm/mach-omap2/omap_hwmod.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod.c b/arch/arm/mach-omap2/omap_hwmod.c index 84cc0bd..f396c83 100644 --- a/arch/arm/mach-omap2/omap_hwmod.c +++ b/arch/arm/mach-omap2/omap_hwmod.c @@ -749,7 +749,7 @@ static int _count_mpu_irqs(struct omap_hwmod *oh) ohii = &oh->mpu_irqs[i++]; } while (ohii->irq != -1); - return i; + return i-1; } /** @@ -772,7 +772,7 @@ static int _count_sdma_reqs(struct omap_hwmod *oh) ohdi = &oh->sdma_reqs[i++]; } while (ohdi->dma_req != -1); - return i; + return i-1; } /** @@ -795,7 +795,7 @@ static int _count_ocp_if_addr_spaces(struct omap_hwmod_ocp_if *os) mem = &os->addr[i++]; } while (mem->pa_start != mem->pa_end); - return i; + return i-1; } /** -- 1.7.4.1
>From c55f38512e1471b8a9405c2cb45d9377e0c45999 Mon Sep 17 00:00:00 2001 From: sricharan <r.sricharan@xxxxxx> Date: Wed, 21 Sep 2011 16:08:46 +0530 Subject: [PATCH 1/8] OMAP: hwmod: Fix the addr space, irq, dma count APIs The address spaces, irqs and dma reqs count API returns the number of corresponding entries in a hwmod including a additional null value or a -1 terminator in the structure introduced recently. More information here: - 212738a4 (omap_hwmod: use a terminator record with omap_hwmod_mpu_irqs arrays) - 78183f3f (omap_hwmod: use a null structure record to terminate omap_hwmod_addr_space arrays) - bc614958 (omap_hwmod: use a terminator record with omap_hwmod_dma_info arrays) The issue with irqs and dma info was originally reported by Benoit Cousson. The devices which have multiple hwmods and use device_build_ss are broken with this, as their resources are populated with a extra null value, subsequently the probe fails. So fix the API not to include the array terminator in the count. Reported-by: Benoit Cousson <b-cousson@xxxxxx> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@xxxxxx> Signed-off-by: sricharan <r.sricharan@xxxxxx> Acked: Benoit Cousson <b-cousson@xxxxxx> --- arch/arm/mach-omap2/omap_hwmod.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod.c b/arch/arm/mach-omap2/omap_hwmod.c index 84cc0bd..f396c83 100644 --- a/arch/arm/mach-omap2/omap_hwmod.c +++ b/arch/arm/mach-omap2/omap_hwmod.c @@ -749,7 +749,7 @@ static int _count_mpu_irqs(struct omap_hwmod *oh) ohii = &oh->mpu_irqs[i++]; } while (ohii->irq != -1); - return i; + return i-1; } /** @@ -772,7 +772,7 @@ static int _count_sdma_reqs(struct omap_hwmod *oh) ohdi = &oh->sdma_reqs[i++]; } while (ohdi->dma_req != -1); - return i; + return i-1; } /** @@ -795,7 +795,7 @@ static int _count_ocp_if_addr_spaces(struct omap_hwmod_ocp_if *os) mem = &os->addr[i++]; } while (mem->pa_start != mem->pa_end); - return i; + return i-1; } /** -- 1.7.4.1
