[PATCHv2 4/4] ARM: mvebu: implement L2/PCIe deadlock workaround
Thomas Petazzoni
thomas.petazzoni at free-electrons.com
Tue May 13 05:52:05 PDT 2014
Dear Arnd Bergmann,
On Tue, 13 May 2014 13:13:39 +0200, Arnd Bergmann wrote:
> Hmm, I think this needs some more explanation about which flags
> you are actually interested in.
>
> These are the three common mem types for ioremap:
>
> #define PROT_PTE_DEVICE L_PTE_PRESENT|L_PTE_YOUNG|L_PTE_DIRTY|L_PTE_XN
> #define PROT_SECT_DEVICE PMD_TYPE_SECT|PMD_SECT_AP_WRITE
>
> [MT_DEVICE] = { /* Strongly ordered / ARMv6 shared device */
> .prot_pte = PROT_PTE_DEVICE | L_PTE_MT_DEV_SHARED |
> L_PTE_SHARED,
> .prot_pte_s2 = s2_policy(PROT_PTE_S2_DEVICE) |
> s2_policy(L_PTE_S2_MT_DEV_SHARED) |
> L_PTE_SHARED,
> .prot_sect = PROT_SECT_DEVICE | PMD_SECT_S,
> .domain = DOMAIN_IO,
> },
> [MT_DEVICE_CACHED] = { /* ioremap_cached */
> .prot_pte = PROT_PTE_DEVICE | L_PTE_MT_DEV_CACHED,
> .prot_sect = PROT_SECT_DEVICE | PMD_SECT_WB,
> .domain = DOMAIN_IO,
> },
> [MT_DEVICE_WC] = { /* ioremap_wc */
> .prot_pte = PROT_PTE_DEVICE | L_PTE_MT_DEV_WC,
> .prot_sect = PROT_SECT_DEVICE,
> .domain = DOMAIN_IO,
> },
>
> and this is the one you enforce here:
>
> [MT_MEMORY_RW_SO] = {
> .prot_pte = L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_DIRTY |
> L_PTE_MT_UNCACHED | L_PTE_XN,
> .prot_l1 = PMD_TYPE_TABLE,
> .prot_sect = PMD_TYPE_SECT | PMD_SECT_AP_WRITE | PMD_SECT_S |
> PMD_SECT_UNCACHED | PMD_SECT_XN,
> .domain = DOMAIN_KERNEL,
> },
>
> So you set a different domain, and turn write-combined and cached mappings
> into uncached mappings, and for uncached mappings you remove the "shared"
> flag. Which of these changes is the one you actually need?
I *believe* the important part is the change from L_PTE_MT_DEV_SHARED
to L_PTE_MT_UNCACHED, because:
#define L_PTE_MT_UNCACHED (_AT(pteval_t, 0x00) << 2)
#define L_PTE_MT_DEV_SHARED (_AT(pteval_t, 0x04) << 2)
So the former is "Strongly-Ordered" according to the ARM ARM, while the
latter is "Device Shareable".
The only detail I have access to is that the workaround is "Reads
targeting PCIe End Point must be marked Strongly Ordered", so it's
pretty limited in details.
Do you think I should create a different memory type MT_DEVICE_SO, that
remains in the DOMAIN_IO domain, but uses L_PTE_MT_UNCACHED instead of
L_PTE_MT_DEV_SHARED ?
Thanks,
Thomas
--
Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
More information about the linux-arm-kernel
mailing list