|
From: | Paolo Bonzini |
Subject: | Re: [RFC PATCH 4/8] qtest/fuzz: Restrict CPU I/O instructions |
Date: | Wed, 17 Mar 2021 18:47:01 +0100 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.7.0 |
On 15/03/21 06:14, Thomas Huth wrote:
diff --git a/tests/qtest/fuzz/generic_fuzz.c b/tests/qtest/fuzz/generic_fuzz.cindex ee8c17a04c4..3e0089f4a63 100644 --- a/tests/qtest/fuzz/generic_fuzz.c +++ b/tests/qtest/fuzz/generic_fuzz.c@@ -304,6 +304,13 @@ static bool get_io_address(address_range *result, AddressSpace *as,return cb_info.found; } +static bool get_mmio_address(address_range *result, + uint8_t index, uint32_t offset) +{ + return get_io_address(result, &address_space_memory, index, offset); +} + +#ifdef TARGET_HAS_IOPORTSorry, but the qtests are generic code, I don't think we should introduce target specific ifdefs here...?
FWIW this is not a qtest, it's a separate emulator executable and this file is compiled per-target.
That said, your objection does apply to patch 5 since libqos is compiled only once for all targets.
Paolo
[Prev in Thread] | Current Thread | [Next in Thread] |