qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-arm <qemu-arm@nongnu.org>, QEMU Developers <qemu-devel@nongnu.org>
Subject: Re: [PATCH v3 04/18] accel/tcg: Add probe_access_flags
Date: Mon, 27 Apr 2020 09:00:23 -0700	[thread overview]
Message-ID: <69d6cc05-2610-d81b-df8f-0f8d723530af@linaro.org> (raw)
In-Reply-To: <CAFEAcA_ugvJWmN8fNbvZYJvOtyv6uPAumQ8UPtK-k9nqN=uerQ@mail.gmail.com>

On 4/27/20 3:48 AM, Peter Maydell wrote:
> probe_access() handles watchpoints. Why doesn't probe_access_flags()
> have to do that?

Because we are explicitly deferring that work to the caller.  That's a good
fraction of the point of the new interface.

>> +        /* Handle clean RAM pages.  */
>> +        if (flags & TLB_NOTDIRTY) {
>> +            notdirty_write(env_cpu(env), addr, 1, iotlbentry, retaddr);
>> +        }
>> +
>> +        /* Handle watchpoints.  */
>> +        if (flags & TLB_WATCHPOINT) {
>> +            int wp_access = (access_type == MMU_DATA_STORE
>> +                             ? BP_MEM_WRITE : BP_MEM_READ);
>> +            cpu_check_watchpoint(env_cpu(env), addr, size,
>> +                                 iotlbentry->attrs, wp_access, retaddr);
>> +        }
> 
> The old code checked for watchpoints first, and then handled notdirty-writes,
> which seems like the more correct order. Why has the new
> version switched them around?

Not an intentional change, but I shouldn't think it would matter in the end.

> The probe_access_internal() doc comment doesn't say that it
> guarantees to set host to NULL for the TLB_MMIO/TLB_INVALID_MASK
> cases, but we implicitly rely on it here.

Eh?  probe_access_internal doesn't have a doc comment.  Call that a bug if you
like, but you seem to be talking about something else.

>> +void *probe_access(CPUArchState *env, target_ulong addr, int size,
>> +                   MMUAccessType access_type, int mmu_idx, uintptr_t retaddr)
>> +{
>> +    void *host;
>> +
>> +    g_assert(-(addr | TARGET_PAGE_MASK) >= size);
>> +    probe_access_flags(env, addr, access_type, mmu_idx, false, &host, retaddr);
>> +    return host;
> 
> The old code returned NULL for a zero size; the new version does not.

Granted.

> The old code passed size into cc->tlb_fill; the new version does not.
> The old code passed size into page_check_range(); the new version does not.

This is the user-only version, and size is not used for tlb_fill.  It is only
trivially used in page_change_range; we have just verified that addr+size does
not cross a page boundary.


r~


  reply	other threads:[~2020-04-27 16:04 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-22  4:32 [PATCH v3 00/18] target/arm: sve load/store improvements Richard Henderson
2020-04-22  4:32 ` [PATCH v3 01/18] exec: Add block comments for watchpoint routines Richard Henderson
2020-04-27  9:27   ` Peter Maydell
2020-04-22  4:32 ` [PATCH v3 02/18] exec: Fix cpu_watchpoint_address_matches address length Richard Henderson
2020-04-27  9:28   ` Peter Maydell
2020-04-22  4:32 ` [PATCH v3 03/18] accel/tcg: Add block comment for probe_access Richard Henderson
2020-04-22  4:32 ` [PATCH v3 04/18] accel/tcg: Add probe_access_flags Richard Henderson
2020-04-27 10:48   ` Peter Maydell
2020-04-27 16:00     ` Richard Henderson [this message]
2020-05-04  9:39       ` Peter Maydell
2020-04-22  4:32 ` [PATCH v3 05/18] accel/tcg: Add endian-specific cpu_{ld, st}* operations Richard Henderson
2020-04-27  9:46   ` Peter Maydell
2020-04-22  4:32 ` [PATCH v3 06/18] target/arm: Use cpu_*_data_ra for sve_ldst_tlb_fn Richard Henderson
2020-04-27 10:51   ` Peter Maydell
2020-04-22  4:32 ` [PATCH v3 07/18] target/arm: Drop manual handling of set/clear_helper_retaddr Richard Henderson
2020-04-22  4:32 ` [PATCH v3 08/18] target/arm: Add sve infrastructure for page lookup Richard Henderson
2020-04-27 11:00   ` Peter Maydell
2020-04-22  4:33 ` [PATCH v3 09/18] target/arm: Adjust interface of sve_ld1_host_fn Richard Henderson
2020-04-22  4:33 ` [PATCH v3 10/18] target/arm: Use SVEContLdSt in sve_ld1_r Richard Henderson
2020-04-22  4:33 ` [PATCH v3 11/18] target/arm: Handle watchpoints " Richard Henderson
2020-04-22  4:33 ` [PATCH v3 12/18] target/arm: Use SVEContLdSt for multi-register contiguous loads Richard Henderson
2020-04-22  4:33 ` [PATCH v3 13/18] target/arm: Update contiguous first-fault and no-fault loads Richard Henderson
2020-04-27 11:03   ` Peter Maydell
2020-04-27 16:16     ` Richard Henderson
2020-04-27 16:32       ` Peter Maydell
2020-04-27 16:45         ` Richard Henderson
2020-04-27 18:38           ` Peter Maydell
2020-04-28 15:02             ` Richard Henderson
2020-04-22  4:33 ` [PATCH v3 14/18] target/arm: Use SVEContLdSt for contiguous stores Richard Henderson
2020-04-22  4:33 ` [PATCH v3 15/18] target/arm: Reuse sve_probe_page for gather first-fault loads Richard Henderson
2020-04-22  4:33 ` [PATCH v3 16/18] target/arm: Reuse sve_probe_page for scatter stores Richard Henderson
2020-04-22  4:33 ` [PATCH v3 17/18] target/arm: Reuse sve_probe_page for gather loads Richard Henderson
2020-04-22  4:33 ` [PATCH v3 18/18] target/arm: Remove sve_memopidx Richard Henderson
2020-04-22  5:37 ` [PATCH v3 00/18] target/arm: sve load/store improvements no-reply

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=69d6cc05-2610-d81b-df8f-0f8d723530af@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).