mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Andrey Smirnov <andrew.smirnov@gmail.com>
To: barebox@lists.infradead.org
Cc: Andrey Smirnov <andrew.smirnov@gmail.com>
Subject: [PATCH 23/58] PCI: designware: Exchange viewport of `MEMORYs' and `CFGs/IOs'
Date: Wed, 12 Dec 2018 23:11:09 -0800	[thread overview]
Message-ID: <20181213071144.31691-24-andrew.smirnov@gmail.com> (raw)
In-Reply-To: <20181213071144.31691-1-andrew.smirnov@gmail.com>

Port of a Linux commit 68a0bfec72cb4f117198ae31df114dad4c5e405d

  When we have only two view ports in a DesignWare PCIe platform, iatu0
  is used for both CFG and IO accesses.  When CFGs are sent to peripherals
  (e.g., lspci), iatu0 frequently switches between CFG and IO.

  For such scenarios, a MEMORY might be sent as an IOs by mistake.
  Considering the following configurations:

    MEMORY  ->   BASE_ADDR: 0xb4100000, LIMIT: 0xb4100FFF, TYPE=mem
    CFG     ->   BASE_ADDR: 0xb4000000, LIMIT: 0xb4000FFF, TYPE=cfg
    IO      ->   BASE_ADDR: 0xFFFFFFFF, LIMIT: 0xFFFFFFFE, TYPE=io

  Suppose PCIe has just completed a CFG access.  To switch back to IO, it
  sets the BASE_ADDR to 0xFFFFFFFF, LIMIT 0xFFFFFFFE and TYPE to IO.  When
  another CFG comes, the BASE_ADDR is set to 0xb4000000 to switch to CFG.  At
  this moment, a MEMORY access shows up, since it matches with iatu0 (due to
  0xb4000000 <= MEMORY BASE_ADDR <= MEMORY LIMIT <= 0xFFFFFFF), it is treated
  as an IO access by mistake, then sent to perpheral.

  This patch fixes the problem by exchanging the assignments of `MEMORYs' and
  `CFGs/IOs', which assigning MEMORYs to iatu0, CFGs and IOs to iatu1.

  We can still have issues with IO transfer, however memory transfer is used
  predominantly therefore we are just minimizing the risk of failure.
  Actually, we can not do much when we have only two viewports.  We can
  either not allow the less frequent IO transfers at all, or can live with a
  remote possibility of getting it corrupted.

  Signed-off-by: Dong Bo <dongbo4@huawei.com>
  [pratyush.anand@gmail.com: Modified commit log to capture remote risk]
  Signed-off-by: Pratyush Anand <pratyush.anand@gmail.com>
  Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>

Signed-off-by: Andrey Smirnov <andrew.smirnov@gmail.com>
---
 drivers/pci/pcie-designware.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/pci/pcie-designware.c b/drivers/pci/pcie-designware.c
index cf1b3032e..817d70850 100644
--- a/drivers/pci/pcie-designware.c
+++ b/drivers/pci/pcie-designware.c
@@ -377,12 +377,12 @@ static int dw_pcie_rd_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 		va_cfg_base = pp->va_cfg1_base;
 	}
 
-	dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX0,
+	dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX1,
 				  type, cpu_addr,
 				  busdev, cfg_size);
 	ret = dw_pcie_cfg_read(va_cfg_base + where, size, val);
 	if (pp->num_viewport <= 2)
-		dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX0,
+		dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX1,
 					  PCIE_ATU_TYPE_IO, pp->io_mod_base,
 					  pp->io_bus_addr, pp->io_size);
 
@@ -416,12 +416,12 @@ static int dw_pcie_wr_other_conf(struct pcie_port *pp, struct pci_bus *bus,
 		va_cfg_base = pp->va_cfg1_base;
 	}
 
-	dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX0,
+	dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX1,
 				  type, cpu_addr,
 				  busdev, cfg_size);
 	ret = dw_pcie_cfg_write(va_cfg_base + where, size, val);
 	if (pp->num_viewport <= 2)
-		dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX0,
+		dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX1,
 					  PCIE_ATU_TYPE_IO, pp->io_mod_base,
 					  pp->io_bus_addr, pp->io_size);
 
@@ -572,7 +572,7 @@ void dw_pcie_setup_rc(struct pcie_port *pp)
         * we should not program the ATU here.
         */
 	if (!pp->ops->rd_other_conf) {
-		dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX1,
+		dw_pcie_prog_outbound_atu(pp, PCIE_ATU_REGION_INDEX0,
 					  PCIE_ATU_TYPE_MEM, pp->mem_mod_base,
 					  pp->mem_bus_addr, pp->mem_size);
 		if (pp->num_viewport <= 2)
-- 
2.19.1


_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

  parent reply	other threads:[~2018-12-13  7:12 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-13  7:10 [PATCH 00/58] PCI i.MX6/DesignWare sync up with 4.20-rc1 Andrey Smirnov
2018-12-13  7:10 ` [PATCH 01/58] PCI: desginware: Remove bogus prototypes Andrey Smirnov
2018-12-13  7:10 ` [PATCH 02/58] PCI: designware: Consolidate outbound iATU programming functions Andrey Smirnov
2018-12-13  7:10 ` [PATCH 03/58] PCI: designware: Use iATU0 for cfg and IO, iATU1 for MEM Andrey Smirnov
2018-12-13  7:10 ` [PATCH 04/58] PCI: designware: Fix PORT_LOGIC_LINK_WIDTH_MASK Andrey Smirnov
2018-12-13  7:10 ` [PATCH 05/58] PCI: designware: Use exact access size in dw_pcie_cfg_read() Andrey Smirnov
2018-12-13  7:10 ` [PATCH 06/58] PCI: designware: Simplify dw_pcie_cfg_read/write() interfaces Andrey Smirnov
2018-12-13  7:10 ` [PATCH 07/58] PCI: designware: Require config accesses to be naturally aligned Andrey Smirnov
2018-12-13  7:10 ` [PATCH 08/58] PCI: designware: Make "num-lanes" an optional DT property Andrey Smirnov
2018-12-13  7:10 ` [PATCH 09/58] PCI: designware: Ensure ATU is enabled before IO/conf space accesses Andrey Smirnov
2018-12-13  7:10 ` [PATCH 10/58] PCI: designware: Simplify control flow Andrey Smirnov
2018-12-13  7:10 ` [PATCH 11/58] PCI: designware: Make config accessor override checking symmetric Andrey Smirnov
2018-12-13  7:10 ` [PATCH 12/58] PCI: designware: Explain why we don't program ATU for some platforms Andrey Smirnov
2018-12-13  7:10 ` [PATCH 13/58] PCI: imx6: Move link up check into imx6_pcie_wait_for_link() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 14/58] PCI: designware: Add generic dw_pcie_wait_for_link() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 15/58] PCI: designware: Add default link up check if sub-driver doesn't override Andrey Smirnov
2018-12-13  7:11 ` [PATCH 16/58] PCI: designware: Move Root Complex setup code to dw_pcie_setup_rc() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 17/58] PCI: designware: Remove incorrect RC memory base/limit configuration Andrey Smirnov
2018-12-13  7:11 ` [PATCH 18/58] PCI: designware: Return data directly from dw_pcie_readl_rc() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 19/58] PCI: designware: Move link wait definitions to .c file Andrey Smirnov
2018-12-13  7:11 ` [PATCH 20/58] PCI: designware: Wait for iATU enable Andrey Smirnov
2018-12-13  7:11 ` [PATCH 21/58] PCI: designware: Check LTSSM training bit before deciding link is up Andrey Smirnov
2018-12-13  7:11 ` [PATCH 22/58] PCI: designware: Keep viewport fixed for IO transaction if num_viewport > 2 Andrey Smirnov
2018-12-14  6:49   ` Andrey Smirnov
2018-12-13  7:11 ` Andrey Smirnov [this message]
2018-12-13  7:11 ` [PATCH 24/58] PCI: designware: Rename dw_pcie_valid_config() to dw_pcie_valid_device() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 25/58] PCI: designware: Simplify pcie_host_ops.readl_rc() and .writel_rc() interfaces Andrey Smirnov
2018-12-13  7:11 ` [PATCH 26/58] PCI: designware: Swap order of dw_pcie_writel_rc() reg/val arguments Andrey Smirnov
2018-12-13  7:11 ` [PATCH 27/58] PCI: designware: Export dw_pcie_readl_rc(), dw_pcie_writel_rc() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 28/58] PCI: dwc: designware: Move register defines to designware header file Andrey Smirnov
2018-12-13  7:11 ` [PATCH 29/58] PCI: dwc: all: Rename cfg_read/cfg_write to read/write Andrey Smirnov
2018-12-13  7:11 ` [PATCH 30/58] PCI: dwc: designware: Get device pointer at the start of dw_pcie_host_init() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 31/58] PCI: imx6: Add local struct device pointers Andrey Smirnov
2018-12-13  7:11 ` [PATCH 32/58] PCI: imx6: Removed unused struct imx6_pcie.mem_base Andrey Smirnov
2018-12-13  7:11 ` [PATCH 33/58] PCI: imx6: Pass struct imx6_pcie to PHY accessors Andrey Smirnov
2018-12-13  7:11 ` [PATCH 34/58] PCI: imx6: Pass device-specific struct to internal functions Andrey Smirnov
2018-12-13  7:11 ` [PATCH 35/58] PCI: imx6: Use generic DesignWare accessors Andrey Smirnov
2018-12-13  7:11 ` [PATCH 36/58] PCI: imx6: Reorder struct imx6_pcie Andrey Smirnov
2018-12-13  7:11 ` [PATCH 37/58] PCI: imx6: Port error messages for imx6_pcie_deassert_core_reset() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 38/58] PCI: imx6: Remove unused return values Andrey Smirnov
2018-12-13  7:11 ` [PATCH 39/58] PCI: imx6: Factor out ref clock enable Andrey Smirnov
2018-12-13  7:11 ` [PATCH 40/58] PCI: imx6: Add DT property for link gen, default to Gen1 Andrey Smirnov
2018-12-13  7:11 ` [PATCH 41/58] PCI: imx6: Remove redundant "Link never came up" message Andrey Smirnov
2018-12-13  7:11 ` [PATCH 42/58] PCI: imx6: Remove LTSSM disable workaround Andrey Smirnov
2018-12-13  7:11 ` [PATCH 43/58] PCI: dwc: all: Split struct pcie_port into host-only and core structures Andrey Smirnov
2018-12-13  7:11 ` [PATCH 44/58] PCI: dwc: designware: Parse "num-lanes" property in dw_pcie_setup_rc() Andrey Smirnov
2018-12-13  7:11 ` [PATCH 45/58] PCI: dwc: designware: Fix style errors in pcie-designware.c Andrey Smirnov
2018-12-13  7:11 ` [PATCH 46/58] PCI: dwc: Split pcie-designware.c into host and core files Andrey Smirnov
2018-12-13  7:11 ` [PATCH 47/58] PCI: dwc: all: Modify dbi accessors to take dbi_base as argument Andrey Smirnov
2018-12-13  7:11 ` [PATCH 48/58] PCI: dwc: all: Modify dbi accessors to access data of 4/2/1 bytes Andrey Smirnov
2018-12-13  7:11 ` [PATCH 49/58] PCI: dwc: designware: Test PCIE_ATU_ENABLE bit specifically Andrey Smirnov
2018-12-13  7:11 ` [PATCH 50/58] PCI: Fix typos and whitespace errors Andrey Smirnov
2018-12-13  7:11 ` [PATCH 51/58] PCI: Add SPDX GPL-2.0 to replace GPL v2 boilerplate Andrey Smirnov
2018-12-13  7:11 ` [PATCH 52/58] PCI: dwc: Replace lower into upper case characters Andrey Smirnov
2018-12-13  7:11 ` [PATCH 53/58] PCI: dwc: designware: Handle ->host_init() failures Andrey Smirnov
2018-12-13  7:11 ` [PATCH 54/58] PCI: dwc: Add accessors for write permission of DBI read-only registers Andrey Smirnov
2018-12-13  7:11 ` [PATCH 55/58] PCI: dwc: Enable write permission for Class Code, Interrupt Pin updates Andrey Smirnov
2018-12-13  7:11 ` [PATCH 56/58] PCI: dwc: Fix enumeration end when reaching root subordinate Andrey Smirnov
2018-12-13  7:11 ` [PATCH 57/58] PCI: dwc: Small computation improvement Andrey Smirnov
2018-12-13  7:11 ` [PATCH 58/58] PCI: dwc: Constify dw_pcie_host_ops structures Andrey Smirnov
2018-12-14  8:28 ` [PATCH 00/58] PCI i.MX6/DesignWare sync up with 4.20-rc1 Andrey Smirnov

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20181213071144.31691-24-andrew.smirnov@gmail.com \
    --to=andrew.smirnov@gmail.com \
    --cc=barebox@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox